Home

Knowledge Modelling in WebOnto and OCML A User Guide

1. i e asserting maps to Meta 1 Hence if we now ask for the mapping between parameters and YQT members we get the following results Knowledge Modelling in OCML Page 37 ask maps to z x t Solution MAPS TO YQT PARAMETER EVA_ EVA_1 Solution MAPS TO YQT PARAMETER MONI KA_X MONI KA_X Solution MAPS TO YQT PARAMETER ULRIKE_U ULRI KE_U Solution MAPS TO YQT PARAMETER UWE_T UWE_T Solution MAPS TO YQT PARAMETER J OACHI M_I J OACHI M_1 Solution MAPS TO YQT PARAMETER HANS W HANS W Solution MAPS TO YQT PARAMETER MICHAEL_T MICHAEL_T Solution MAPS TO YQT PARAMETER ANGY_ W ANGY_W Solution MAPS TO YQT PARAMETER J URGEN_L J URGEN_L Solution MAPS TO YQT PARAMETER KATHARI NA_N KATHARI NA_N Solution MAPS TO YQT PARAMETER THOMAS _D THOMAS D Solution MAPS TO YQT PARAMETER HARRY_C HARRY _C Solution MAPS TO YQT PARAMETER ANDY _L ANDY_L Solution MAPS TO YQT PARAMETER MARC_M MARC_M Solution MAPS TO YQT PARAMETER WERNER _L WERNER_L The advantage of this solution is that parameters and YQT members maintain their separate identities as shown in the next box Knowledge Modelling in OCML Page 38 describe instance YQT PARAMETER WERNER _L Instance YQT PARAMETER WERNER_L of class YQT PARAMETER HAS VALUE RANGE C5 123 C5 122 C5 121 C5 120 C5 119 C5 117 C5 116 C5 113 C5 114 5
2. 6 KNOWN BUGS Currently there are three known bugs which can not be fixed trivially 1 Deselection of items Whenever an item is selected in either the item list graphical area or the fine grained view window selected items in the other windows are deselected Unfortunately it is not possible to detect a selection using a single mouse click in the item list You can select an item in the item list using a single mouse click but the event is not sent to WebOnto Consequently when an item is selected in the item list the currently selected items in the other two windows are not deselected 2 Double clicking Double clicking on an item in the item list or the fine grained view window causes the item to be inspected Double clicking is not always detected This varies between hardware and operating system platforms 3 No warnings before deleting a class When deleting a class which has instances there is no warning on the feature that every instance of the class is also deleted Chapter 2 Knowledge Modelling in OCML Enrico Motta The modelling language described in this chapter OCML was originally developed in the context of the VITAL project Shadbolt et al 1993 to provide operational modelling capabilities for the VITAL workbench Domingue et al 1993 Over the years the language has undergone a number of changes and improvements and in what follows I will present an overview of the current version of the language v6 3 and
3. if a new class cl ass1 is linked to an existing class cI ass2 the source The WebOnto User Guide Page 14 code defclass class1 class2 is written The editing of instances is aided by the automatic creation of forms These are described in more detail in section 5 2 Prompting for component parts in figure 15 the user has linked the class academic to the class ur When a user does this a selector wor ks at in figure 15 is automatically created The selector contains all the slots within the academic class so the user has to merely select a slot rather than recall and type a slot s name Semantic colour coding all the links drawn by the user between graphical icons are automatically colour coded depending on the type of link e g class to subclass links are dark green and class to instance links are blue 5 2 Scenario We shall describe the features available in WebOnto s edit mode using a short scenario In the scenario below we describe how a user adds some classes and instances to the tutorial ontology ontology The user selects the ontology and initiates an editing session in using the following steps The user l 2 Selects Clear Diagram Before Drawing under the Edit menu Selects the Graph Entire Ontology Tree menu item under the Ontology menu see figure 11 Each knowledge model within the displayed tree is colour coded according to the knowledge model type 1 Beige the basic ont
4. 115 describe instance WERNER _L Instance WERNER_L of class RESEARCHER HAS PROJECT RESPECT SMOKER NO HACKER YES WORKS WITH ANGY W MARC_M GROUP YQT Formally a mapping can be characterized as an association between an object say o and its meta object m o so that the entity denoted by the entity denoted by m o is the same as the entity denoted by o Genesereth and Nilsson 1988 This notion can be formalized in OCML by means of the following definition def relation maps to x y This relation allows the user to specify an association between an object at the task layer and one at the domain layer Formally y denotes the object denoted by the object denoted by x no op iff def y denotation x 4 2 Relation mapping Instance mapping works only in those cases in which imposing a view over a domain can be reduced to creating task level mirror images for a finite number of domain level objects A more general scenario is one in which there is some relation defined at the task method level which needs to be reflected to the domain level in a dynamic fashion A well known example is that in which domain concepts or statements are viewed as hypotheses at the problem solving level This association is typically dynamic given that hypotheses are considered as such only for a particular time slice of the problem solving process These situations can be modelled in OCML by means of relation
5. 1988 Appendix 1 Additional Details on OCML 1 FUNCTIONAL TERM CONSTRUCTORS A BNF specification of each term constructor is provided as well as an informal description of its operational semantics Setofal setofal finds all solutions i e environments to basic log expression and then returns the list obtained by instantiating template in all the returned environments ensuring that the list contains no duplicates If no solutions are found then the empty list i e ni is returned setofall term setofall template basic log expression template nil term term term constant variable string fun term findall term l the term l in env term l quote term l if term cond term fun the name of a function or a term constructor constant A symbol whose first character is not variable A symbol whose first character is string A lisp String e g string log expression quant log expression basic log expression quant log expression foral schema or var log expression exists schema or var log expression basic log expression and log expression l or log expression l gt log expression log expression lt gt log expression log expression not log expression l rel expression Appendix 1 Additional Details on OCML Page 49 schema or var schema variable schema variable schema ni rel expression rel term rel a
6. compare it to alternative knowledge modelling languages Moreover I will also illustrate a subset of the application development interface supporting the construction of OCML models This interface consists of a number of Lisp macros and functions which can be used for retrieving and modifying OCML constructs for evaluating functional and control terms and for querying an OCML model 1 TYPES OF CONSTRUCTS IN OCML OCML supports the specification of three types of constructs functional and control terms and logical expressions 1 1 Functional terms A functional term in short a term specifies an object in the current domain of investigation A functional term can be a constant a variable a string a function application or can be constructed by means of a special term constructor This can be one of the following if cond the setofall findall quote andin environment see appendix 1 for a description of the semantics of these terms Variables are represented as Lisp symbols beginning with a question mark e g x is a variable Strings are sequences of characters enclosed in double quotes e g string A function application is a term such as fun fun term where fun is the name of a function and fun term a functional term Functions are defined by means of the Lisp macro def f unction which is described in section 2 2 Functional terms are evaluated by means of the OCML function interpreter 1 2 Control terms Model
7. description Specifically in accordance with a view of knowledge as a competence like notion Newell 1982 it is useful to decouple the level at which we describe what an agent knows from the level at which we describe how the agent organizes and infers knowledge Such an approach is used for instance in the Cyc system Lenat and Guha 1990 which integrates multiple knowledge representation techniques the heuristic level but provides a uniform interface to the Cyc knowledge base the epistemological level A similar approach is also followed by Levesque 1984 which describes a logic based query language which can be used to communicate with a knowledge base at a functional level independently from the data and inference structures present in the knowledge base In particular in the case of OCML this generic idea of providing a uniform level of description to a hybrid formalism has been instantiated by providing a Tell Ask interface Levesque 1984 which use logical expressions i e a relation oriented view when modifying or querying an OCML model independently of whether the query in question concerns a class a slot or a ordinary relation The key to this integrated view is the fact that classes and slots are themselves relations classes are unary relations and slots are binary ones In addition to supporting a generic Tell Ask interface this property makes it possible to provide rich specifications of classes and slo
8. fun The option constraint provides a way to constrain the domain i e the set of possible inputs of a function It specifies a logical expression which must be satisfied by the input arguments of the function The def option indicates a logical expression which defines the function This expression should be predicated over both some input arguments and the output variable Operationally the expression denoted by the constraint option provides a mechanism for testing the feasibility of applying a function to a set of arguments The expression denoted by def provides a mechanism for verifying that the output produced by a function application is consistent with the formal definition of the function Finally the options body and lisp fun provide effective mechanisms for computing the value of a function The former specify a functional term which is evaluated in an environment in which the variables in the function schema are bound to the actual arguments The latter makes it possible to evaluate an OCML function by means of a procedural attachment expressed as a Lisp function The arity of this Lisp function should be the same as that of the associated OCML function def function filter 1 rel gt sub Returns all the elements in 1 which satisfy rel def and unary relation rel list 1 list sub 1 gt and member x sub 1 holds rel x member x 1 body if null 1 if hold
9. informally describes their relevance to constraint checking and theorem proving As shown in the table constraint checking is supported by the following keywords constraint def and iff def While these have different model theoretic semantics see table 2 1 from a constraint checking point of view they are equivalent They all specify an expression which has to be satisfied by each known instance of the relevant relation The relation options iff def sufficient prove by and lisp fun provide Relation Supports Provides Option constraint checking proof mechanism iff def constraint def Table 2 2 Operationally relevant relation options in OCML Knowledge Modelling in OCML Page 26 mechanisms for verifying whether or not a relation holds for some arguments The first two iff def and sufficient also play a specification role see table 2 1 The others prove by and lisp fun only play an operational role Both iff def and sufficient indicate logical expressions which can be used to prove whether some tuple is an instance of a relation From a theorem proving point of view there is an important difference between them Let s suppose we are trying to prove that a tuple say T satisfies a relation say R Ifa sufficient condition is tried and failed the OCML proof system will then search for alternative ways of proving that T satisfies R If an iff def condition is tried and failed then no alternative pr
10. m 1 if professor domain reference m 71 setofall r double a type room r usable yes def rule has value range 2 has value range gen m 1 i f secretary domain reference m 71 setofall r and room r size n usable yes gt n 1 close to r kmi entrance 2 8 2 Forward rules OCML also allows the user to define forward rules A forward rule comprises zero or more antecedents and one or more consequents Antecedents are restricted to relation expressions while any logical expression can be a consequent When a forward rule is executed OCML treats each consequent as a goal to be proven and attempts to prove them until one fails This mechanism makes it possible to integrate data driven and goal driven reasoning and to specify arbitrarily complex right hand sides A special operator exec is provided to allow OCML users to introduce control and therefore functional terms in the right hand side of a rule In particular two useful procedures are tel to assert new facts and out put to produce output A simple example showing how to use these in a forward chaining rule is given below Knowledge Modelling in OCML Page 35 def rule foo has project x y then exec tell project covered by y x exec output has project S S x y While forward rules can be useful in a number of situations when building application models e g to define watchers which are triggered
11. mappings A relation mapping provides a mechanism to associate rules and procedures to a relation say R so that when a query of type R is posed or assertions of type R are made at the task method level these events can be reflected to the domain level The purpose of these reflection actions is to ensure that the consistency between domain and task method levels is maintained An example of an upward relation mapping is illustrated by the definition below which is taken from an application model developed for the Sisyphus I problem The mapping is an upward one in the sense that it is used fo lift van Harmelen and Balder 1992 the office allocation statements existing at the domain level to the problem solving level Knowledge Modelling in OCML Page 39 Specifically the goal of this mapping is to associate the relation current design model which is used by the parametric design problem solver to indicate the design model associated with the current design state to the set of i n room assertions present in the current snapshot of the domain knowledge base The relation maps to is used to retrieve the parameter associated with each particular YQT member def relation mapping current design model up current design model dm if dm setofall p v and in room x v maps to p x An upward relation mapping ensures that when a task method level relation is needed the relevant information is obtaine
12. symbol naming a relation Findal findall is the same as set ofal except that it does not remove duplicate solutions findall term findall template basic log expression The the finds one solution i e environment to basic log expression and then returns the list obtained by instantiating template in the returned environment If no solutions are found then the constant nothing is returned the term the template basic log expression In environment The primitive in environment takes a list possibly empty of pairs var term and a body and returns the result of evaluating this in an environment in which each var is bound to termi in env term in environment pairs body pairs nil pair pairs pair variable term body i term Quote The value of an expression such as quote term is term quote term term quote term If The first action which is carried out when evaluating an if term is to check whether log expression is satisfied If this is the case then then term is evaluated in the environment which satisfies log expression If log expression cannot be satisfied in the current model then there are two possibilities If else term is specified then this is evaluated and the value obtained is returned as the value of the if term If else term is not present and log expression cannot be proved then the constant not hing is returned Appendix 1 Additional Details on OCML
13. those inherited from more generic ones see appendix 1 for more details on the inheritance mechanism in OCML 2 4 OCML instances Instances are simply members of a class An instance is defined by means of def instance which takes as arguments the name of the instance the parent of the instance i e the most specific class the instance belongs to optional documentation and a number of slot value pairs An example of instance definition taken from the Sisyphus I domain model is shown in the box below def instance harry_c researcher has project babylon smoker no hacker yes works with jurgen_ thomas d In particular the above definition shows that a slot can have multiple values In this case harry_c works both withj urgen_ andthomas_d 2 5 Object oriented and relation oriented approaches to modelling When describing classes and instances I made use of standard object oriented terminology and talked about slots having values and instances belonging to classes This object centred approach is in a sense orthogonal to the relation centred one which I used when discussing relations and logical expressions The former focuses on the entities populating a model and then associates properties to them the latter centres on the type of relations which characterize a domain and then uses these to make statements about the world These two approaches to modelling representation have complementary strengths and weaknesses
14. whenever some situation arises in a knowledge base they are not essential to the model building process The reason for this is that knowledge level modelling is mainly about constructing definitions while forward chaining rules are about behaviour Thus they can be used in place of procedures to describe behaviour but they cannot replace constructs for relation or function specification 3 FUNCTIONAL VIEW OF OCML A functional view of a knowledge representation system focuses on the services the system provides to the user Levesque 1984 Brachman et al 1985 Basically there are three kinds of services provided by OCML 1 operations for extending modifying a model ii interpreters for functional and control terms and iii a proof system for answering queries about a model Extensive details on the model extension modification facilities provided by OCML were given in earlier sections 4 MAPPING OCML is meant to support both the specification of library components and the development of partial or complete application models according to the Task Method Domain Application TMDA framework described in Motta 1999 The role of the TMDA framework is to support application development by reuse and to this purpose it distinguishes between domain generic tasks and problem solving methods on one side and domain specific models on the other Applications are created by instantiating generic problem solving components tasks and methods i
15. 1 for a description of WebOnto s edit mode 2 3 Object Draw draw the current item in the graphical display area Draw Tree draw a tree in the graphical display area if appropriate for the currently selected item Currently trees are drawn for classes tasks problem solving methods and knowledge models Inspect inspect the currently selected object displaying the result in the fine grained view window Direct Instances show the direct instances of the currently selected class in the fine grained view window All Instances show all the instances of the currently selected class in the fine grained view window The WebOnto User Guide Page 6 Display Direct Instances show the direct instances of the currently selected class in the graphical display area Display All Instances show all the instances of the currently selected class in the graphical display area Draw Tree Vertically if selected trees are drawn vertically rather than horizontally 2 4 Ontology View Classes display all the classes in the current knowledge model in the Item list View Top_classes display the classes with no parents in the current knowledge model in the Item list View Relations display all the relations in the current knowledge model in the Item list View Functions display all the functions in the current knowledge model in the Item list View Procedures display all the procedures in the current knowledge
16. 220799 knowledge base e The menu bar we describe the menu items in the following section e The tool bar the tool bar icons are described in section 2 8 e The graphical display area Graphical representations of selected objects from the Item list can be displayed here The WebOnto User Guide Page 2 KNOWLEDGE ME INSTIFUTE Welcome to WebOnto constructed within the HCREMA Project This system was implemented by John Domingue with input from Enrico Motta Mare Risenstadt and Zdenek Zdrahal You can Jind some useful documents such as the user guide at the webonto home page Operations Annotation Broadcast Figure 2 An annotated screen snapshot of WebOnto in browsing mode depicting the applet components The WebOnto User Guide fea Clears i PEE E R a E E T BME eo 1 kmi inmtleilie enrichel intranet k Mome Ontology amp ol snitulogy Decne nbation Te choology belong wsed Eon Eocich Proserts Sapee Cleese wib agqeac cechsolagy fet nudel ling technology web based eck Lao Thherit i Sot Taimi dint lal en bape ducektion Ftari time TYE tins point Eimas ches poine deer au ble C PE Peceo ahbg wee Ho Ledge oedia iaie Epe orgenisation Seeds pote oer Spe fheeccomre bleactorn Hina op oe Loe ye bear CAPE OPSTSAC i yb Pate Of Re reg ee F bEpEi pUbtware technology ot abe GPR gon tua ce ooecwe Pechrotogy baitdr an a lues meh tipe becker ogy wire rary hea apei km
17. Broadcast aai goal specification 4 EID a Tj Java Applet Window Figure 8 A screen snapshot showing a user selecting a number of nodes Has baw plaan kh TRE datk H Figure 9 A screen snapshot showing 3 selected nodes These nodes can be moved or cut in a single operation if desired The WebOnto User Guide Page 13 5 1 Edit Mode Figure 10 A screen snapshot showing how the appearance of WebOnto changes when it is in edit mode Figure 10 shows how the appearance of WebOnto changes when it is put into edit mode Two new icons appear in the tool bar a Edit the currently selected object and Create an instance of the currently selected class Additionally a well from which new OCML definitions can be created is displayed The graphical display area can be used to create OCML definitions by direct manipulation WebOnto understands the semantics of OCML and prevents the user from creating semantically incorrect definitions in four ways 1 Node linking highlighting Nodes can only be linked if this action is semantically correct As the user moves the mouse around linkable nodes are highlighted Node highlighting can be turned using the Node Highlighting menu item under the Edit menu 29 2 Automatic creation of templates and forms The manipulation of new nodes e g linking them to existing nodes results in the automatically insertion of source code For example
18. CML language Schreiber et al 1994 which supports ontological specifications and the construction of Common KADS models CML supports the definition of various constructs including concepts attributes tasks methods relations structures and expressions Obviously the main difference between CML and OCML is that the former is only meant to be an informal notation while the latter is a fully operational language Another important difference is that CML is committed to supporting the Common Knowledge Modelling in OCML Page 43 KADS framework while the kernel of the OCML language is framework independent In addition to the KADS related commitments CML also embodies other modelling commitments it provides primitives for representing structures and part of relations This approach has both advantages and disadvantages On the plus side it extends the range of modelling primitives provided by the language and supports notions which occur frequently in conceptual modelling On the other hand there are two possible problems with this approach The first one has to do with embedding ontological commitments in a conceptual modelling language In particular different approaches to modelling structure and aggregation can be found in the literature e g compare the analysis by Martin and Odell 1995 with that by Lenat and Guha 1990 Embedding one particular approach in the kernel of a conceptual modelling language prevents users from extending the la
19. ICAL EXPRESSIONS iaooe an eer pe ieee en aeaa AE ENA E hiss aides EE E ENE EE REE ir 24 Basic Domain Modelling in OCML 24 2 1 OCME RELATIONS okan n ea e ea iat ae ae ee en ce i 24 2 1 1 Relation specification options cccecccsceecceeceesesseensensecsecusecsecuseesecuseeeceseeseseaeeseeeaeeeasenaengs 24 2 1 2 Operationally relevant relation OPtiONs scccsccsseesecssceseceeceseeseceseeeeceseeeeseseeeeseseesaeeasents 25 2 1 3 A meta option for non operational SPeCIfICATIONS cscceseccesecerseceeeenseceaceesneeeseceenecnseeeaees 26 2 1 4 OCML relations SUMMING Upsista ais Ees NEE E e eneen 27 2 2 OCML FUNCTIONS ea eretas oe Er obs cee EEE ETE tas cs anete obits EEEE E E EE E EEEE EEEE N iY 28 2 3 OCIVIE GIEASSES Ee 29 2 4 OCM INSTAN OES E n test cobs betes E E E E EE EEE EEEE E E EERE E 30 2 5 OBJECT ORIENTED AND RELATION ORIENTED APPROACHES TO MODELLING ssseeeesseeesrsrese see 30 2 6 THE GENERIC TELL ASK INTERFACE sccssscsscesecsecescesecesceseceseesncesesaecaeseaecaassaecnaesascnseeseeseaeees 31 2 6 1 Tell a generic ASSertiON MAKING primitive cececesecsscesscesceseeeeceseeeeceseeseceseeeeseseeeaeensente 31 2 6 2 Ask a generic Query POSING primitive eeeecesceeeeseenceneeeacensecaecesecaecuaeeaecuseeecneeeseeeeeeaeees 32 2 7 OCME PROCEDURES ore ori u eroare nO E EErEE EE EIEEE EEE EE ENEE SEEE EES E EEE EEE EEEE SENS 33 2 8 RULE BASED REASONING IN OCML seseeeeeeeseeeesereresrereresesseressreresterereretesesee
20. KNOWLEDGE MEIA INSTITUTE Knowledge Modelling in WebOnto and OCML A User Guide John Domingue Enrico Motta and Oscar Corcho Garcia Version 2 4 Table of contents Introduction The WebOnto User Guide A Ww 6 Overview 1 Menu and tool Bars 4 2 1 BIER cs cecods ENE OESTE EE EEEE AE OE EAEE EEEE OS AEE E E E EE 4 22 FED ED Sesece covsceecsdetdescustavessnscceade ES ESE EEE EE EEEE EEE E EEEE eS 5 2 3 OBIEC Tiss oes seecsdetadescetdavdsastoe cede ciedeceads eben EE O EEEE EEE EE EEEE 5 2 4 ONTOLOGY eed conc de S EEPE ASE ENE OEE oda seein oa gtusn E EEE E 6 2 5 OPERATIONS EENE EOE EEE O EEEE EEEE EEEE OR 7 2 6 ANNOTATIONS 3 924 esses ocssssocdadeoss ASE EE OEE EEE E EEEE 7 Dof IBROADCAST AROEN AAE ESEE EO A EEE E EE EEEE A 7 2 8 TOOLBAR EOTS EEE EEE EOE EEEE EN 8 Creating and Deleting Knowledge models 8 Item List 10 Graphical Display area 10 5 1 J EBI UMAY LO B Sa eae Se oo SP EEEE A A A E EE 13 5 2 SCENARIO AEA EEE AEREA I OE A EAEE E EAE AE A E E E E ANT 14 5 3 SUMMARY OF MOUSE GESTURES IN THE GRAPHICAL AREA sssccccesessssececessaeceececssteceseeeeenees 21 known bugs 22 Knowledge Modelling in OCML 1 2 Types of Constructs in OCML 23 1 1 FUNCTIONAL TERMS 3 5 s2essccsscssssaspascasgeesteasssscoecessaassoesdsh sess adds ous sas cede hacesuasdest TE 23 1 2 CONTROL TERMS 3 ss355s siscssssaspdstascesgitachassovcoeceas ys ssoesesbubsssasdsous bas cebechacheges adestaeyesscsbeatseasseesbsastade 23 1 3 LOG
21. Miami Florida February 1991 Martin J and Odell J J 1995 Object Oriented Methods A Foundation Prentice Hall Englewood Cliffs New Jersey Motta E Reusable Components for Knowledge Models Principles and Case Studies in Parametric Design IOS Press 1999 Newell A 1982 The knowledge level Artificial Intelligence 18 1 pp 87 127 Riva A and Ramoni M 1996 LispWeb a Specialised HTTP Server for Distributed AI Applications Computer Networks and ISDN Systems 28 7 11 1996 953 961 Schreiber A Th Wielinga B J Akkermans H van de Velde W and Anjewierden A 1994 CML The CommonKADS Conceptual Modelling Language In L Steels A T Schreiber and W van de Velde Editors A Future for Knowledge Acquisition Proceedings of the 8th European Knowledge Acquisition Workshop Springer Verlag LNAI 867 pp 283 300 Shadbolt N Motta E and Rouge A 1993 Constructing Knowledge Based Systems IEEE Software 10 6 pp 34 39 November 1993 Shaw M and Garlan D 1996 Software Architectures Perspectives on an Emerging Discipline Prentice Hall 1996 References Page 47 Wiederhold G and Genesereth M 1997 The Conceptual Basis for Mediation Services IEEE Expert September October Issue pp 38 47 Yen J Neches R and MacGregor R 1988 Classification based Programming A Deep Integration of Frames and Rules Technical Report ISI RR 88 213 USC Information Science Institute March
22. Page 50 if term if log expression then term else term then term term else term term Cond The interpreter iterates through each clause of a cond term until it finds one whose log expression is satisfied If none is found then nothing is returned Otherwise let s assume cond clause is the first clause whose log expression is satisfied In this case the value of the cond term is obtained by evaluating the term associated with cond clause cond term cond cond clause cond clause log expression term 2 CONTROL TERM CONSTRUCTORS A BNF specification of each control term constructor is provided as well as an informal description of its operational semantics In environment The primitive in environment takes a list possibly empty of pairs var termy and a control body and returns the result of evaluating this in an environment in which each var is bound to term in env control term in environment pairs control body pairs nil pair pairs pair variable term control body control term control term term l in env control term l if control term l cond control term l do control term l loop control term l repeat control term l return control term if control term if log expression then control term else control term then control term control term else control term control term Appendix 1 Additional Details on OCML Page 51 cond control t
23. a one of the source files for the current knowledge model The user is prompted with the names of the source files on the WebOnto server where the current knowledge model is stored The selected file is displayed in a display source window Debug turn on an internal debugger that sends a trace of all operations to the Java console window Quit quit WebOnto 2 2 Edit Graphical Cut remove the currently selected object from the graphical display area Delete delete the currently selected item only available in edit mode Edit Source edit the currently selected object only available in edit mode View Source display the OCML source of the currently selected object New Class Instance creates an instance definition form see section 5 2 enabling an instance of the selected class to be created only available in edit mode Search for String search for OCML entities within the library which contain the specified string Clear Diagram clear the graphical area Clear Before Drawing if selected the graphical area is always cleared before drawing a new item Node Highlighting turn on node highlighting This feature uses the semantics of OCML to highlight which components are allowed to be connected to each other This prevents users from attempting to create links that have no defined semantics in OCML only available in edit mode Edit Mode when selected WebOnto is placed in edit mode See section 5
24. and for this reason they are often combined in knowledge representation and modelling languages to provide hybrid formalisms Fikes and Kehler 1985 Yen et al 1988 In the context of a knowledge modelling language rather than a knowledge representation one the main advantage gained from combining multiple paradigms is one of flexibility Both object oriented and relation oriented approaches provide conceptual frameworks which make it possible to impose a view over some domain The choice between one or the other can be made for ideological or pragmatic reasons e g whether the target delivery environment is a rule based shell or an object oriented programming environment In the specific context of an operational modelling language such as OCML another benefit which is gained by providing support for both object oriented and relation oriented modelling is that these approaches are naturally associated with particular types of inferences Object orientation provides the structure for inheritance Knowledge Modelling in OCML Page 31 and automatic classification relation orientation is normally associated with constraint based and rule based reasoning While the integration of multiple paradigms provides the aforementioned benefits when describing or interacting with a knowledge model it is useful to abstract from the various modelling paradigms and inference mechanisms integrated in the model and characterize it at a uniform level of
25. appropriate ontology This approach has the advantage of flexibility different modelling frameworks can then be supported through the specification of the relevant ontologies A more recent version of KARL which is called New KARL Angele et al 1996 does away with the strong KADS oriented approach used by KARL and focuses instead on the specification of task method structures and ontology mappings Thus New KARL subscribes to a modelling framework which has much more in common with the OCML task ontology than the one underlying KARL However in contrast with OCML New KARL also hardwires these task and method centred primitives in the language itself rather than in a particular ontology Other formal specification languages exist for KADS models see Fensel and van Harmelen 1994 for an overview While the formal details of these languages of course vary similar conclusions to those drawn above can be reached when comparing them to OCML these languages tend to emphasize formal aspects and are based on a KADS approach OCML emphasizes operationality and flexibility and does not presuppose although it can support a KADS approach Indeed the original raison d tre for OCML was to provide an operational alternative to a formal specification language KgsSF Jonker and Spee 1992 in the context of the VITAL workbench Domingue et al 1993 Among the informal notations available for knowledge modelling the most notable is the
26. cation organization instance form for the slot works at generated automatically by WebOnto The WebOnto User Guide Page 21 Figure 21 A screen snapshot showing WebOnto when the editing session has been terminated 5 3 Summary of Mouse Gestures in the Graphical Area Within WebOnto a number of operations can be carried out using mouse gestures in the Graphical Display window The effect of a mouse gesture is dependent on WebOnto s current mode browse or edit and the gesture s start and end points We summarise the possible operations in the following tables In Browse Mode Gesture Start End Result Mouse Down Node Select the node Control Mouse Down Prompt for a text annotation Mouse Drag Node Move a node Mouse Drag Not over a Node Specify a select region In Browse Mode with the Trail menu item under the Annotation Menu selected Gesture Start End Result Mouse Drag Draw a freehand line The WebOnto User Guide Page 22 In Edit Mode Gesture Start End Result Control Shift Mouse Node Create a text window to Down change the node name Control Mouse Drag Class Node1 Class Create a superclass link Node2 from node1 to node2 Control Mouse Drag Instance Nodel Class Create an instance of link Node2 from node1 to node2 Control Mouse Drag Class Nodel Instance Create a default value link Node2 from node1 to node2 Control delayed Shift Class Nodel Class Create a type of link from Mouse Drag Node2 nodel to node2
27. clausej cond control term cond cond control clause Loop The control construct oop provides a simple mechanism for iterating over lists It first evaluates a term which should return a list say L Then it iterates over each element of L say I and evaluates control term in an environment in which variable is bound to J loop control term loop for variable in term do control term Appendix 1 Additional Details on OCML Page 52 Do The control construct do is a simple sequencing primitive The control terms in its body are evaluated sequentially once only do control term do actions fcontrol term Repeat The control term constructor repeat repeats the control term s specified in its body until the end test is satisfied if the test has the form until test Otherwise if the test has the form while fest then repeat actions stops as soon as the test fails If the end test is specified after the control terms then the control terms are carried out at least once i e the end test is verified at the end of each cycle If the end test is specified before the control terms then the test is verified at the beginning of each cycle If no test is provided then all control expression in the body of arepeat actions are repeated ad infinitum repeat control term repeat actions fend test control term repeat actions fcontrol term fend test Return This is a simple way of exi
28. d from the domain level Of course problem solving is also about inferring knowledge and retracting previously held assertions Hence OCML also supports downward relation mappings These divide into two categories add and remove The former specifies a procedure which is activated when a new relation instance is asserted The latter specifies a procedure which is activated when a relation instance is removed In the case of relation current desi gn model relation mappings are needed to ensure that when the design model considered by the problem solver is modified the relevant changes are reflected onto the domain model see definition below def relation mapping current design model down add lambda x do unassert in room any m any r loop for pair in x do if maps to first pair z tell in room z rest pair Finally the definition below shows the remove downward mapping associated with relation current desi gn model it simply removes the domain level assertions associated with the design model which is passed as argument to the relation instance being retracted def relation mapping current design model down remove lambda x loop for pair in x do if maps to first pair z unassert in room z rest pair Knowledge Modelling in OCML Page 40 5 ONTOLOGIES OCML also provides support for defining ontologies When an ontology is defi
29. d view window after the works at button has been pressed The second column contains a text window into which a value can be typed If the slot is typed a third and fourth column are provided The third column is a menu containing the types defined for the slot and the subclasses of the defined types The fourth column is a menu containing the instances of the selected type 2 Starts to fill in the works at slot Examples of how the works at slot has been used shown in figure 17 are obtained by pressing the works at button Selects higher educational organization from the type menu see figure 18 4 Selects New Instance from the instances menu see figure 19 This causes a new instance form for the class hi gher educational organization to be created see figure 20 5 Fill in the instance form for the hi gher educati onal organi zation class 6 Fill in the remainder of the professor instance slots and hits the OK button Finally the user ends the session by selecting the Edit Mode menu item under the Edit menu WebOnto indicates that the persons and organizations ontology is now unlocked The appearance of WebOnto at the end of the session is shown in figure 21 The WebOnto User Guide Page 17 Prien el E if Figure 12 A screen snapshot showing a new class node C ass1 being dragged from the well E3 Editing Class Class1 def class Classl academic Figure 13 A sc
30. dow see figure 13 to senior_lecturer and hits the Done button The user then specifies that the has web address slot of the academic class should be of type ur by carrying out the following steps The user 1 Defines a new class ur in the following manner The user drags the Cl ass node from the well onto the graphical display area clicks on the a icon edits the code in the edit window changing the class name from Class2 to url and then clicks on the Done button 2 Clicks on the class academic with the mouse and drags to the class url The user presses and holds down the Shift button and then releases the mouse button see figure 14 The WebOnto User Guide Page 16 3 Clicks on the label unset and chooses has web address from the selector label see figure 15 The definition of the academi c class is automatically changed to def class academic educational employee has web address type url The user creates an instance of professor using the following steps 1 Selects the professor node in the graphical display area and clicks on the 2 icon The form shown in figure 16 appears Each slot in an instance form is represented by a row containing two or four columns The first column contains a button with the slot name Pressing the button causes all examples of the use of the slot to be displayed in the fine grained view window Figure 17 shows the output in the fine graine
31. en the Trail menu item under the Annotation menu see section 2 6 is selected dragging the mouse with the left button pressed draws a line in the selected Trail Colour Clicking in the graphical area with the control button pressed creates a prompt for a text annotation note that WebOnto must be in browsing mode In figure 7 the user has drawn a red trail around a subclass link and then created a text annotation When the Trail button is not selected then dragging with the left mouse button draws a shaded area enabling multiple nodes to be selected This is shown in figures 8 and 9 below Single node selection is shown by four pimples one at each corner of the node Multiple node selection as shown in figure 9 is shown by dark shading The WebOnto User Guide fg User oscar Browsing kmi planet kb drafts 220799 classes File Edit Object Sv av educational emplo educational organ J educational resear electronic publishit email address empty set event event involving m event involving pr event involving pr exam taking place executable task extemally funded p female child function gender general medicine generalized mover generalized transte generic agent genetic algorithms geographical regic geographically dis giving a lecture giving a tutorial Ontology Operations event involving production sodal gathamg C e Annotation Page 12
32. ent Stanford University References Page 46 Genesereth M R and Nilsson N J 1988 Logical Foundations of Artificial Intelligence Morgan Kaufmann Los Altos CA Gruber T R 1993 A Translation Approach to Portable Ontology Specifications Knowledge Acquisition 5 2 Guarino N 1994 The Ontological Level In R Casati B Smith and G White Editors Philosophy and the Cognitive Science Holder Pichler Tempsky Vienna Austria van Harmelen F and Balder J R 1992 ML 2 A Formal Language for KADS Models of Expertiese Knowledge Acquisition 4 1 pp 127 161 Jonker W and Spee J W 1992 Yet Another Formalization of KADS Conceptual Models In Th Wetter K D Althoff J Boose B R Gaines M Linster and F Schmalhofer Editors Current Developments in Knowledge Acquisition EKAW 92 pp 112 32 LNAI 599 Springer Verlag Berlin Lenat D B and Guha R V 1990 Building Large Knowledge Based Systems Representation and Inference in the Cyc Project Addison Wesley Reading MA Levesque H J 1984 Foundations of a functional approach to knowledge representation Artificial Intelligence 23 2 pp 155 212 Linster M 1994 Problem Statement for Sisyphus Models of Problem Solving International Journal of Human Computer Studies 40 2 pp 187 192 MacGregor R 1991 Using a Description Classifier to Enhance Deductive Inference Proceedings of the 7th IEEE Conference on AI Applications
33. erm cond cond control clause t cond control clause log expression control term loop control term loop for variable in term do control term do control term do actions control term repeat control term repeat actions fend test control term repeat actions fcontrol term fend test end test i while test until test test log expression return control term return term f The first action which is carried out when evaluating an if control term is to check whether log expression is satisfied If this is the case then then control term is evaluated in the environment which satisfies log expression If log expression cannot be satisfied in the current model then there are two possibilities If else control term is specified then this is evaluated and the value obtained is returned as the value of the if control term If else control term is not present and log expression cannot be proved then the constant not hi ng is returned if control term if log expression then control term felse control term Cond The interpreter iterates through each clause of a cond control term until it finds one whose log expression is satisfied If none is found then nothing is returned Otherwise let s assume cond clausej is the first clause whose log expression is satisfied In this case the value of the cond control term is obtained by evaluating the control term associated with cond
34. es a kind of environmental support for the construction of OCML models It includes special operators like exec which makes it possible to invoke procedures from rules and procedures such as out put which prints out a message Task Method This ontology provides the concepts required to model tasks and problem solving methods i e to support the construction of task and problem solving models This set of ontologies provides a rich modelling platform from which to build other ontologies and or problem solving models It is natural to compare the OCML and Ontolingua base ontologies Farquhar et al 1996 There are two aspects which distinguish these two sets of ontologies their nature and their scope The first difference is related to the operational nature of OCML The Ontolingua base ontology is not concerned with operationality and therefore includes many non operational definitions The OCML base ontology is concerned with providing support for the construction of operational models As a result it attempts to minimize the number of non executable specifications A typical approach is to weaken a non operational definition to make it executable For instance let s consider the function universe In the Ontolingua base ontology the universe of a relation is defined as the set of all objects for which the relation is true Of course this is not an operational definition However we can provide a weaker version of universe called k
35. essions The bullet points below describe how non basic goal expressions are proven in OCML Appendix 1 Additional Details on OCML Page 55 and A B This expression is satisfied if both A and B can be proven in the current model or AB This expression is satisfied if either A or B can be proven in the current model gt A B This expression is satisfied if either A cannot be proven or if B can be proven in each environment which is a solution to A lt gt A B This expression is satisfied if both gt A B and gt BA can be proven not A This expression is satisfied if A cannot be proven in the current model exists schema or var A This expression is satisfied if A can be proven in the current model forall schema or var gt A B This expression is satisfied if either A cannot be proven or if B can be proven in each environment which is a solution to A Thus the proof mechanism supported by OCML is not complete with respect to first order logic statements In particular disjunctions can only be proved by proving each clause separately and negated expressions are only proved by default
36. f selected dragging with the mouse button pressed draws a trail that is a freehand line Trail Colour choose the colour to trail in Clear clear the currently displayed trails and text annotations 2 7 Broadcast Broadcast when selected WebOnto broadcasts all the user s interface actions to WebOnto clients in receive mode The WebOnto User Guide Page 8 e Receive when selected all the broadcast interface actions are received by the current WebOnto Receiving WebOnto clients are in effect slaves of the broadcasting WebOnto 2 8 Tool Bar Below the menu bar there is a tool bar containing the following icons Draw the selected OCML component in the graphical display area B Create a graphical representation for the currently selected OCML component If a class is selected a class hierarchy is drawn In figure 2 above the class hierarchy for the kmi technology class is displayed oY Inspect the currently selected OCML component The results are displayed in the fine grained view window In figure 3 above the kmi technol ogy class is currently being inspected So View the last item that was inspected The fine grained view window keeps a history of all the items inspected This icon enables the user to move back through the history View the next item in the fine grained view window history View the OCML source of selected item H Search for a string in the library of knowledge
37. g the class name person would cause only instances of person to be displayed Delete Filter delete a defined previously filter Select Ontology reads in the OCML structures in the selected knowledge model and displays all the classes in the item list It should be noted that this is subtly different from the View items above View selects between different OCML types within a knowledge model This item allows the user to switch knowledge models It is recommended that this item is used when changing knowledge models Unlock Ontology when a knowledge model is edited the WebOnto server locks it so that no one else can edit it at the same time Unfortunately if WebOnto crashes whilst a knowledge model is being edited the knowledge model can remain locked This item can be used to unlock the knowledge model Only the root user can do this Graph Entire Ontology Tree draw a graph of the inheritance hierarchy of all the knowledge models in library 2 5 Operations Tell prompt for an OCML expression to invoke with the OCML tell operation Ask prompt for an OCML expression to invoke with the OCML ask operation Set of All prompt for an OCML expression to invoke with the OCML setofall operation Unassert prompt for an OCML expression to unassert in the current knowledge model Evaluate prompt for an OCML expression to invoke with the OCML evaluate operation 2 6 Annotations Trail i
38. h depend on pfun If term has the format fun termo term with n 0 where fun is the name of a function and a Lisp body associated with fun exists then ocml eval returns the value obtained by applying the Lisp body to the values obtained by evaluating each term in env If term has the format fun termo term with n 0 where fun is the name of a function and no Lisp body associated with fun exists then ocml eval returns the value obtained by applying the body of fun to the values obtained by evaluating each term in env In all other cases ocmi eval signals an error The OCML interpreter for control terms Control terms are interpreted in a manner analogous to functional terms The control term interpreter is implemented by a Lisp macro ocml control eval which has the following behaviour If term is a functional term then it is evaluated according to the rules given in section 4 1 If term has the format proc termo term with n 0 where proc is a primitive control operator then term is evaluated in env according to criteria which depend on proc If term has the format proc termo term With n 0 where proc is the name of a procedure and a Lisp body associated with proc exists then oc ml control eval returns the value obtained by applying the Lisp body to the values obtained by evaluating each term in env If term has the format proc termo termn With n 0 where proc is the
39. he clauses in the order in which these are listed in the rule definition As in Prolog depth first search with chronological backtracking is used to control the proof process I use the following notational conventions when describing the syntax of OCML constructs Braces and are used to indicate that the enclosed item is optional For instance the notation x means that x may or may not be present but if it is present it can only appear once The notation x means that x can appear 1 or more times i e it must appear at least once while the notation x indicates that x can appear 0 or more times Square brackets within braces are used in situations in which a number of alternatives for a non terminal item are possible but each alternative can be used only once For example let s consider a non terminal item x for which alternatives a and b are possible In this context the notation x y indicates that any number of xy sequences are possible but a or b can only appear once in practice this means that we can have between 0 and 2 sequences Braces can also be nested For instance the notation x y means that both x and y are optional but y can only appear if x does Finally I use italics to denote non terminal items and a vertical bar to indicate mutually exclusive alternatives Knowledge Modelling in OCML Page 34 Both semantically and operationally a backward chaining rule is the same as a sufficient
40. i rceoecenccbh themrn Citeamces fedh 6 Ched 1 Leet D popde 0c EAEE cerh gn Ohad Llnteenet pope diap eleccroum is tech snri ched 1ntesost experience srcchivs ce ch enkichedl La Leenet Cein workbook eloton Tiina oom Scheer foals Page 3 Figure 3 A snapshot of the fine grained view window currently showing the class kmi knowledge enriched intranet Figure 3 shows the fine grained view window Messages from the server and the internal descriptions of OCML components displayed using the il icon are displayed here Any text displayed in the fine grained view window is automatically parsed and any word that corresponds to an OCML item is displayed in a specific colour The colour of a word indicates the OCML type The current colour coding is e Class green e Instance turquoise The WebOnto User Guide Page 4 e Relation blue e Function violet e Procedure violet red e Rule maroon e Ontology orange Highlighted items can be selected and operated on In figure 3 the relation end time is currently selected Double clicking on an item has the same effect as selecting it and clicking on the of inspect icon Items can be selected in the Item List Graphical Display Area or the Fine Grained View window Selecting an item in one area deselects any selected items in the other areas but see known bug 1 in section 6 2 MENU AND TOOL BARS We shall now describe the menu bar The me
41. igms frames rules message passing it integrates the various constructs according to a classification centred viewpoint While this approach obtains nice results in terms of inferential capabilities it nevertheless can be a constraint for the knowledge analyst who is forced to frame the current problem within a classification centred framework In contrast with this approach OCML provides a number of alternative modelling constructs e g rules functions classes and procedures which while integrated are themselves primitive modelling components and can be used within different modelling approaches The notion of mapping presented in this chapter plays a role similar to that of mediators in heterogeneous information systems Wiederhold and Genesereth 1997 connectors in software architectures Shaw and Garlan 1996 and adapters in design patterns Gamma et al 1995 The idea underlying all of these approaches is essentially the same some kind of external kit is required in order to allow the interaction of reusable components and their configuration for different computational scenarios The externalization of this adaptation process has the advantage that the original components remain unchanged while they become usable in the new situation 7 SYNOPSIS The role of the OCML language is to provide operational knowledge modelling facilities To this end it includes interpreters for functional and control terms as well as a pr
42. intersection set partition set cardinality etc Numbers Defines the various concepts and operations required for reasoning about numbers and for performing calculations Lists Defines the concepts and operations associated with lists It includes classes such as list and atom functions such as first rest and append and relations such as member Strings Specifies the concepts and operations associated with strings e g string string append etc Mapping This ontology defines the concepts associated with the mapping mechanism described earlier It includes only three definitions relation maps to and functions meta reference and domain reference The former takes a domain level instance and returns the associated task method level instance The latter performs the inverse function Frames Defines the concepts associated with the frame based representation used in OCML It comprise meta classes such as cl ass and instance functions such as direct instances and all slot values relations such as has one and has at most and procedures such as append slot value Inferences The purpose of this ontology is to provide a repository for defining functions and relations supporting the specification of KADS like inferences So far Knowledge Modelling in OCML Page 41 only a few such inferences have been added to this ontology to support different types of selection and sorting Environment This ontology provid
43. l simply by asking whether a logical statement is satisfied The OCML proof system will then carry out the relevant inference and retrieval operations depending on whether the relation being queried is a slot a class or an ordinary relation The process is however transparent to the user For instance we can find out about the projects in which harry_c is involved after the assertion shown above these are now babylon and mit by using the Lisp macro ask to pose the query has project harry_c c The resulting interaction with the OCML proof system is shown below ask has project harry_c c Solution HAS PRO ECT HARRY _C BABYLON More solutions y or n y Solution HAS PROJECT HARRY _C MLT More solutions y or n y No more solutions This uniform relation centred view over OCML models also provides a way to index inferences For instance when answering the above query the OCML proof system will first retrieve and order all inference mechanisms applicable to a query of type has project e g these might include assertions of type has proj ect relation options associated with has proj ect and the relevant backward rules and will then try these in sequence to generate one or more solutions to the query more details on the OCML proof system are given in appendix 1 7 See section 2 8 1 for a description of OCML backward rules Knowledge Modelling in OCML Page 33 2 7 OCML proced
44. l of description and I will illustrate the various primitives which are provided in OCML to support the specification of logical expressions functional and control terms In particular OCML provides mechanisms for defining relations functions classes instances rules and procedures 2 1 OCML relations Relations allow the OCML user to define labelled n ary relationships between OCML entities Relations are defined by means of a Lisp macro def relati on which takes as arguments the name of a relation its argument schema optional documentation and a number of relation options An argument schema is a list possibly empty of variables Relation options play two roles one related to the formal semantics of a relation the other to the operational nature of OCML These roles are discussed in the next two sections 2 1 1 Relation specification options From a formal semantics point of view the purpose of a relation option is to help to characterize the extension of a relation Table 2 1 shows the relation options which can be used to provide formal relation specifications and for each option informally describes its semantics A formal semantics to these options can be given in terms of the homonymous Ontolingua constructs 2 The careful reader will have noticed that I have already mentioned if and cond when discussing functional terms In fact if and cond can be used to construct both functional and control terms However this does no
45. ling problem solving behaviour involves more than making statements and describing entities in the world Control terms are needed to specify actions and describe the order in which these are executed OCML supports the specification of sequential iterative and conditional control structures by means of a number of control term In what follows I will uset his font to refer to expressions in the OCML language Knowledge Modelling in OCML Page 24 constructors such as repeat loop do if and cond A Lisp macro def procedure makes it possible to label parametrized control terms i e to define procedures Control terms are evaluated by means of a control interpreter 1 3 Logical expressions OCML also provides the usual machinery for specifying logical expressions The simplest kind of logical expression is a relation expression which has the form rel fun term where rel is the name of a relation and fun term is a functional term More complex expressions can be constructed by using the logical operators and or not gt lt gt and quantifiers forall and exists Operational semantics is provided for all operators and quantifiers Relations are defined by means of the Lisp macro def relation which is described in section 2 1 2 BASIC DOMAIN MODELLING IN OCML In the previous section I have introduced the three types of constructs which are supported by OCML In this section I will go down at a more fine grained leve
46. louse and Caylus France September pp 320 339 Springer Verlag Domingue J 1998 Tadzebao and WebOnto Discussing Browsing and Editing Ontologies on the Web Proceedings of the 11th Banff Knowledge Acquisition Workshop Banff Alberta Canada April 18 23 1998 Farquhar A Fikes R and Rice J 1996 The Ontolingua Server A Tool for Collaborative Ontology Construction In B Gaines and M Musen Editors Proceedings of the 10th Banff Knowledge Acquisition for Knowledge Based Systems Workshop Banff Alberta Canada Fensel D 1995 The Knowledge Acquisition and Representation Language KARL Kluwer Dordrecht Fensel D and van Harmelen F 1994 A Comparison of Languages which Operationalize and Formalize KADS Models of Expertise The Knowledge Engineering Review 9 2 Fensel D and Straatman R 1996 Problem solving methods Making Assumptions for Efficiency Reasons In N Shadbolt K O Hara and Schreiber G Editors Advances in Knowledge Acquisition EKAW 96 Lecture Notes in Artificial Intelligence 1076 Springer Verlag Heidelberg Fikes R E and Kehler T 1985 The Role of Frame Based Representation in Reasoning Communications of the ACM 28 9 September 1985 Gamma E Helm R Johnson R and Vlissides J 1995 Design Patterns Addison Wesley Genesereth M R and Fikes R E 1992 Knowledge Interchange Format Version 3 0 Technical Report Logic 92 1 Computer Science Departm
47. model in the Item list View Rules display all the rules in the current knowledge model in the item list View Instances display all the instances in the current knowledge model in the item list View Tasks display all the tasks in the current knowledge model in the item list View problem_solving_methods display all the PSMs in the current knowledge model in the item list View All display all the OCML structures in the current knowledge model in the item list View Only Items in Current Ontology if selected OCML entities from inherited knowledge models are not displayed Include Base Ontology if selected OCML entities from the base ontology are displayed Change Number of Viewable Items set a limit on the number of items which can be viewed in the item list This option facilitates the browsing of large knowledge models The default value is 150 Add View Filter this menu item allows the user to create a set of filters in order to filter viewable items in the item list They can be combined using the operators and and or e Add a String Filter show only items containing the specified string e Add a Not String Filter show only items not containing the specified string The WebOnto User Guide Page 7 e Add a Letter Range Filter show only items with their first letter in the specified range e Add a Predicate Filter show only items for which the given predicate is true For example typin
48. models B Edit the source of the selected item only available in Edit Mode B Create new instance of the selected class only available in Edit Mode 3 CREATING AND DELETING KNOWLEDGE MODELS We ll now describe how WebOnto can be used to create delete and edit the properties of a knowledge model Figure 4 shows the dialog created when the menu item New Ontology under the file menu is chosen The dialog is used to enter the four properties for the knowledge model namely 1 Name the name of the knowledge model In figure 4 an ontology called vehi cI e ontology 1s being created 2 Ontology Uses the names of any knowledge models that this knowledge model should use i e inherit from The snapshot in figure 4 specifies that the vehicle ontology will use the engi neering ontology ontology 3 Allowed Editors the names of any registered users or groups each WebOnto user belongs to a group who can edit the knowledge model Figure 4 specifies that angela and any user belonging to the vehicle design group will be allowed to edit the ontology The WebOnto User Guide Page 9 4 Ontology Type the type of the knowledge model this can be one of domain method task or application knowledge base Name vehicle ontology Ontology Uses fenaineering ontology Allowed Editors angela vehicle design Ontology Type Domain 7 Ok Cancel 213 Java Applet Window Figure 4 A screen snapshot of the C
49. models thus moving away from the batch oriented translation based operationalization model suggested for Ontolingua Gruber 1993 The design philosophy underlying the KARL language Fensel 1995 has many points in common with OCML In particular both KARL and OCML are operational modelling languages and are therefore suitable for rapid prototyping of knowledge level models Moreover both KARL and OCML are part of comprehensive knowledge engineering frameworks providing methodological support for KBS development However there are also differences KARL is an executable formal specification language where the emphasis is on formalization its main strength is the provision of a formal semantics for its modelling constructs The design philosophy of OCML has more to do with pragmatic considerations its main goal is to provide a flexible modelling environment able to support different approaches to modelling rapid prototyping and structured development executable and non executable constructs formal and informal specifications The two languages also differ in terms of the modelling frameworks they employ KARL is based on the KADS four layer framework while OCML more precisely the task method ontology is based on the TMDA framework An important difference is also that while the primitives in the KARL language closely reflect the KADS approach the OCML kernel is approach neutral Its commitment to the TMDA framework is defined by means of the
50. n a particular domain Earlier versions of the language provided support for task and method specification by means of special purpose modelling constructs The current version does away with these task and method specific constructs and only provides a basic set of domain modelling facilities the extension to the language required to specify tasks and methods is then defined as a particular representation ontology the task method sub ontology of the OCML base ontology The advantage of this approach is that it separates the core set of logical primitives the OCML kernel from additional framework specific epistemological commitments Thus only special purpose primitives for defining mapping knowledge are required in addition to standard domain modelling capabilities This mapping knowledge support application modelling by allowing the customization of generic components in a particular domain In the next sections I will discuss the two kinds of mapping constructs supported by OCML relation mapping and instance mapping Knowledge Modelling in OCML Page 36 Matiaad Clnaiogy Parmeter Maaping Mechaniam Choma Ginio Empres Figure 2 1 Mapping domain to method ontologies 4 1 Instance mapping Figure 2 1 illustrates a simple example in which a class of concepts at the task method level parameter is mapped to a class of concepts at the domain level employee This is a very common situation when developing systems thr
51. name of a procedure and no Lisp body associated with proc exists then ocml control eval returns the value obtained by applying the body of proc to the values obtained by evaluating each term in env In all other cases ocml control eval signals an error Appendix 1 Additional Details on OCML Page 54 4 2 The OCML proof system 4 2 1 Procedure for proving basic goal expressions in OCML Let s suppose we want to find all solutions to a basic goal expression say G with format rel fun term where rel is the name of a relation and fun term a functional term In general we might be interested in one some or all solutions Therefore the order in which solutions are generated might be important The algorithm used by the OCML proof system is as follows 1 If rel is not a defined relation then signal an error Otherwise initialize SOLI SOL2 SOL3 SOL4 SOLS and SOL6 to the empty set and go to step 2 Retrieve all the assertions present in the current model whose type i e first element is rel Match each assertion with G All successful matches we call this set SOLJ provide solutions to G Go to step 3 If a Lisp attachment exists for rel then evaluate it in the Lisp environment to find eventual additional solutions to G say SOL2 Go to step 8 If a prove by proof condition say prove rel expression has been specified for relation rel then compute all solutions to prove rel expression say SOL3 Each of these is als
52. nd part of the document describes OCML an operational knowledge modelling language which provides the underlying representation for the ontologies and knowledge bases which can be developed using WebOnto This second part is a revised version of chapter 4 of Motta 1999 The third part Appendix 1 gives more details about the interpreters and reasoning facilities provided by OCML This document is available from the WebOnto home page at http kmi open ac uk projects webonto Chapter 1 The WebOnto User Guide John Domingue 1 OVERVIEW You can start WebOnto by going to http webonto open ac uk This page has the appearance of figure 1 The page contains an invisible applet from which the classes for the WebOnto applet are loaded After 20 seconds the applet frames shown in figures 2 and 3 should appear We call these the main WebOnto window and the fine grained view window Figure 2 shows a screen snapshot of the WebOnto client browsing the knowledge base called kmi planet kb drafts 220799 The main window at the top has five areas which are e The title bar this shows the name of the knowledge model currently being browsed and the type of the components currently displayed in the Item List OCML distinguishes between four types of knowledge models domain ontologies problem solving methods task ontologies and knowledge bases The title in figure 2 indicates that we are currently viewing the classes in kmi planet kb drafts
53. ned say O it is possible to specify which ontologies are included in O and as a result O will include all the definitions from its parent ontologies When conflicts are detected e g the same concept is defined in two different parent ontologies a warning is issued Primitives for loading and selecting ontologies are also provided By default all ontologies are built on top of the OCML base ontology This comprises twelve sub ontologies which include the basic definitions required to reason about basic data types e g lists numbers sets and strings the OCML system itself and the OCML frame representation Specifically the following sub ontologies are provided Meta This ontology defines the concepts required to describe the OCML language It includes constructs such as OCML expression functional term rule relation function assertion etc This ontology is particularly important to construct reasoning components which can verify OCML models Functions Defines the concepts associated with functions e g it includes relations such as domain range unary function bi nary function etc Relations Defines the various notions associated with relations These include the universe and the extension of a relation the definition of reflexive and transitive relations partial and total orders etc Sets This ontology defines the notions associated with sets e g empty set union
54. nguage according to an alternative approach The second problem is caused by the provision of different levels of description i e logical epistemological and conceptual Brachman 1979 within the same formalism Much work in knowledge representation over the past twenty years has focused on identifying the different levels at which knowledge representation languages can be specified Brachman 1979 Guarino 1994 In particular the paper by Brachman clearly illustrates that much of the confusion surrounding the field of knowledge representation in the seventies was caused by the fact that researchers were comparing formalisms which were situated at different levels e g logical and conceptual Thus it seems to me that including ontological primitives e g structures in a language characterized at the logical level is a potential source of confusion This is especially the case with informal languages given that eventual ambiguities are not explained by the underlying formal theory The LOOM language MacGregor 1991 is strictly speaking a knowledge representation rather than a knowledge modelling language i e it also includes symbol level representation constructs However its formal kernel i e what are called the terminological and assertional components provides purely logical and epistemological primitives and therefore it can be used for knowledge modelling and ontology specification The main feature of LOOM is its powerful cla
55. nown uni verse which returns the set of all entities which are part of a tuple satisfying the relation in question This function can be either defined separately from uni verse or attached to it to provide an operational definition The second difference concerns the scope of the two base ontologies Both the Ontolingua and OCML base ontologies provide a rich set of definitions for domain modelling In order to comply with the requirements imposed by the TMDA framework the OCML base ontology provides support also for specifying tasks and problem solving methods 6 COMPARISON WITH OTHER LANGUAGES In the previous section I compared the base ontologies provided by OCML and Ontolingua and emphasized that the differences between them have mainly to do with the conceptual requirements imposed by the TMDA framework on OCML and with its operational nature If we compare OCML and Ontolingua purely as modelling languages their main difference has to do with the fact that while Ontolingua is concerned exclusively with ontologies i e term specification OCML aims to model behaviour as well For this reason OCML also provides support for defining control terms Apart from this aspect the current version of OCML closely mirrors many of the constructs in Ontolingua Thus while OCML extends Ontolingua in various respects it is possible although reductive to view OCML as an environment for prototyping Knowledge Modelling in OCML Page 42 Ontolingua
56. nu bar contains the following menu items 2 1 File e Reload Ontology reload an ontology on the server This need only be used when a set of files have been placed on the server by the server administrator All the knowledge models are loaded when the server is started up When import export facilities are added to WebOnto this menu item is expected to be more heavily used e Login login into WebOnto Note that one can only edit create or delete knowledge models when logged in Contact j b domingue open ac uk if you require a login name and password e Logout logout of WebOnto e Change Password change the currently logged in user s password e New Ontology create a new knowledge model This menu item is described in section 3 e Delete Ontology delete a knowledge model This is menu item described in section 3 e Edit Ontology Properties edit the properties of a knowledge model The properties include the author of the knowledge model and the additional users who are allowed to edit the knowledge model This menu item is described in section 3 e Save diagram save the currently displayed items in the display area to a file on the server e Open diagram open a previously saved diagram e Reinitialise OCML clear all the currently defined OCML structures and reload the entire library on the server Only the root user can do this The WebOnto User Guide Page 5 Display Source File display
57. o a solution to G Go to step 8 If a iff def proof condition say iff rel expression has been specified for relation rel then compute all solutions to iff rel expression say SOLA Each of these is also a solution to G Go to step 8 If a sufficient proof condition say suff rel expression has been specified for relation rel then compute all solutions to suff rel expression say SOLS Each of these is also a solution to G Go to step 7 If a backward chaining rule has been specified associated with relation rel then invoke it to find all other solutions to G say SOL6 Go to step 8 The set of all solutions to query G is obtained by appending the lists SOLI SOL2 SOL3 SOL4 SOLS and SOL6 The algorithm shown above provides an operational semantics for the various relation forming constructs provided by OCML In particular the following two points should be highlighted Assertions inherited through an isa hierarchy are always cached at definition time This means that they are retrieved at step 2 when the goal is matched against the current set of known facts The results returned by non logical mechanisms such as Lisp attachments and pr ove by are only merged with the results obtained by simple assertion matching step 2 In other words they are meant to provide efficient proof mechanisms which override those provided by definition forming options such as iff def and sufficient 4 2 2 Proof rules for non basic goal expr
58. of the currently viewed knowledge model The Ontology menu see section 2 4 allows the user to control which items within the knowledge model are displayed Using the Ontology menu the user can specify e That only items of a specified OCML type classes instances relations functions procedures rules PSMs tasks are displayed e Whether OCML entities inherited from ancestor knowledge models are displayed e Whether OCML entities from the base ontology are displayed e An numeric limit on the number of items displayed this is useful when browsing very large ontologies e Filters which are used to cut down the number of items displayed Again this is useful when browsing very large ontologies 5 GRAPHICAL DISPLAY AREA The WebOnto User Guide Page 11 ij eee Gry tee Figure 7 A screen snapshot showing a user annotating part of a class hierarchy in broadcast mode The graphical display area can be used for three different purposes e Knowledge model visualization graphical representations of OCML entities can be drawn in the area using the oS icon e Synchronous communication when WebOnto is in broadcast mode interface actions are copied to all receiving WebOnto clients e Editing OCML definitions can be created and edited using the graphical display area Figure 7 shows a user using the sketching facilities in WebOnto to communicate with a colleague whilst in broadcast mode Wh
59. ology 2 Light gold a domain ontology 3 Light red an application or knowledge base 4 Light blue a problem solving method 5 Yellow a task ontology Selects the node t ut ori al ontology see figure 11 Selects the Select Ontology menu item under the Ontology menu Selects the Edit Mode menu item under the Edit menu The WebOnto User Guide Page 15 Figure 11 A screen snapshot of WebOnto displaying the structure of the knowledge models within the library This can be obtained by selecting the Graph Entire Ontology Tree menu item under the Ontology menu The colours indicate the knowledge model type The user defines a new class seni or_ ecturer which is a subclass of the class academi c by using the following steps Specifically the user f Selects the academic class in the item list and clicks on the amp icon 2 Drags the Cl ass node from the well onto the graphical display area Notice see figure 12 that the node label for the new class Cl ass1 is displayed in a bold italic font because no definition for the node exists on the WebOnto server 3 Clicks on the academic class node in the graphical display area with the mouse whilst holding the control key down and drags a line to the Cl ass1 node 4 Selects the Cl ass1 node and clicks on the 7 icon This creates the edit window shown in figure 13 5 Changes the name of the class in the edit win
60. oof system which integrates inheritance with backward chaining function evaluation and procedural attachments While the emphasis here is on operationality OCML aims to support different styles of knowledge modelling Therefore OCML provides for first order logic definitions and allows the user to explicitly distinguish non operational from operational definitions Moreover it supports an extensive set of Ontolingua constructs and therefore can be used as an interpreter for Ontolingua definitions References Angele J Decker S Perkuhn R and Studer R 1996 Modeling Problem Solving Methods in New KARL In B Gaines and M Musen Editors Proceedings of the 10th Banff Knowledge Acquisition for Knowledge Based Systems Workshop Banff Alberta Canada Brachman R J 1979 On the Epistemological Status of Semantic Networks In N V Findler Editor Associative Networks Representation and Use of Knowledge by Computers Academic Press New York pp 3 50 Brachman R J Fikes R E and Levesque H J 1985 KRYPTON A Functional Approach to Knowledge Representation In R J Brachman and H J Levesque Editors Readings in Knowledge Representation Morgan Kaufmann Los Altos CA Domingue J Motta E and Watt S 1993 The Emerging Vital Workbench In Ed Aussenac N Boy G Gaines B Linster M Ganascia J G and Kodratoff Y Knowledge Acquisition for Knowledge Based Systems 7th European Workshop EKAW 93 Tou
61. oof mechanism will be attempted The relation options prove by and 1 isp fun are meant to support rapid prototyping and early validation by providing efficient mechanisms for checking whether a tuple satisfies arelation The difference between prove by and isp fun has to do with the expressions which are used as values to the two options prove by points to a logical expression i sp fun to a non logical one specifically a Lisp expression The box below provides an example of how the various types of relation options can be used concurrently to specify a relation and to support constraint checking and efficient proofs This example is taken from a task ontology for parametric design problems Motta 1999 see http chaucer open ac uk 3000 webonto ontology parametric design The relation has val ue shown below associates a design parameter to its value in a design model The definition specifies that v is the value of a parameter p in a design model dm if and only if the pair p v is an element of dm In addition it also specifies the constraint that the value of a parameter has to be a member of its value range if this has been specified Finally the definition includes a pr ove by option whose value is an expression which can be used for verifying whether the relation is satisfied for a triple p v dm This expression provides an efficient proof method by weakening the i f f def statement which formall
62. ough reuse a problem solving domain independent model imposes a particular view over a set of domain concepts Fensel and Straatman 1996 A simple case is one in which a domain view is constructed by direct association of task level concepts to domain level concepts For instance a parametric design view over the Sisyphus I domain can be imposed simply by creating parameter instances and associating them to YQT members Thus the set of parameters and the set of YQT members are associated but kept distinct This solution is appealing for two reasons it supports reuse of modular components and does not confuse the two different types of concepts parameters and YQT members maintain different sets of properties and different semantics thus avoiding a situation in which a design parameter has a wife and a YQT member has a value range Instance mapping is supported in OCML by means of the Lisp macro def upward cI ass mapping This takes the names of two classes as arguments and associates each instance of the first class to a purpose built instance of the second class By default the relation maps to is used to associate the task level instance to the domain level one In the Sisyphus I application model we should then state def upward class mapping yqt member yqt parameter The above form iterates over each instance of class yqt member say creating a new instance of class yqt parameter say Meta I and associating this to
63. pecified or it has been specified and it is merge all default values apply If the inheritance facet has been specified and it is supersede then the value of s in I is obtained by i ranking the ancestors of I according to the class precedence order of the parent of I and ii Appendix 1 Additional Details on OCML Page 53 4 retrieving the default value of the first class in the class precedence order which specifies a default value for s The details of the algorithm used to compute the class precedence order are given at pp 782 786 of the Common Lisp specification Steele 1992 This algorithm produces a total order if this exists based on two ordering principles i a class say C precedes all its direct superclasses and ii a direct superclass of c precedes the direct superclasses of C specified to its right in the list of direct superclasses of C INTERPRETERS AND PROOF SYSTEM 4 1 The OCML interpreter for functional terms The OCML interpreter is implemented by means of a Lisp macro ocml eval This evaluates a functional term term in an environment env according to the following rules 1 2 3 10 11 12 13 If term is a variable then the binding of term in env is returned If term is a string or a constant then term is returned If term has the format pfun termo term with n 0 where pfun is a primitive term constructor then term is evaluated in env according to criteria whic
64. reate New Ontology dialog box The Delete Ontology menu item under the file menu allows users to delete any knowledge models that they own The dialogue box displayed for this menu item displays any knowledge models that the current WebOnto user owns not the ones that s he is just allowed to edit except for the current knowledge model When a user selects the Edit Ontology Properties menu item under the file menu a dialog containing a menu of all the knowledge models owned by the user is displayed A snapshot of this dialog is shown in figure 5 When the user selects the OK button the dialog shown in figure 6 is displayed This dialog is similar to the dialog for creating new knowledge models The only difference is that the dialog allows the author of the knowledge model to be changed FE Edit the Properties of an Ontology Ontology kmi planet kb drafts 220799 4 oK gl Java Applet window Figure 5 A screen snapshot of the first Edit Ontology Properties dialog box The menu on the right contains all the knowledge models owned by the current user The WebOnto User Guide Page 10 S Edit the Properties of kmi planet kb PIE E3 Name kmiplanetkb drafts 220793 Allowed Editors fohnen Ontology Type Domin H OK Cancel Tja j Java Applet Window Figure 6 A screen snapshot of the second Edit Ontology Properties dialog box 4 ITEM LIST The item list displays items
65. reen snapshot showing the edit window created by selecting the Class1 node shown in figure 9 and clicking on the a icon The WebOnto User Guide Page 18 Figure 14 A screen snapshot just after the user has created a type link from the academic Class to the ur class Figure 15 A screen snapshot just after the user has clicked on the unset label in the type link in figure 14 and before selecting has web address The WebOnto User Guide Page 19 paa Instance of professor ss eS E address email address None 0 E organization Figure 16 A screen snapshot of the pr of essor instance form automatically created by WebOnto f Information on the slot works at amples of the use of the slot works at here are no examples of the instances of the same class Examples with instances other classes DANIEL had the value THE OPEN UNIVERSITY Figure 17 A screen snapshot of the fine grained view window showing examples of the use of the slot works at This message was obtained by pressing on the works at button in the instance form show in figure 16 above Figure 18 A screen snapshot of type menu for the wor ks at slot of the professor instance form The WebOnto User Guide Page 20 Figure 19 A screen snapshot of the instance menu for the type hi gher educati onal organization for the work at slot of the professor instance form Figure 20 A screen snapshot of the hi gher edu
66. relation option they both provide an expression which is sufficient to verify that a tuple holds for a relation Thus one might wonder whether rules are needed at all In practice the advantage of including backward rules in the language is that these provide a modular mechanism for refining existing possibly generic relation specifications for instance in cases where application specific knowledge is needed to complete the specification of a relation To clarify this point let s consider an example taken from the KMi office allocation problem This was a real world problem that our institute faced when moving to a new building back in 1997 The relation has val ue range is defined in the parametric design task ontology to characterize the set of possible values which can be assigned to a design parameter When building an application model the generic has val ue range specification is usually refined so that the space of possible values that can be assigned to a particular parameter is precisely defined A modular way to achieve this is to refine the definition of the relation by means of the appropriate backward chaining rules The rules shown below fulfil this purpose for two of the classes of parameters present in the KMI office allocation domain professors and secretaries In this example the former can only go into a double room the latter into a large room next to the entrance def rule has value range 1 has value range gen
67. resseetesterereeetesesreseset 33 2 8 1 Backward EE E ese ee 33 2 8 2 POFWOATA a AEE EEE ences coat E EEEE OE O O EEE AT EEEE 34 3 Functional View of OCML 4 Mapping 4 1 INSTANCE MAP N a a a e a e eaaa 4 2 RELATION MAPPING a a R A E r eae A 5 Ontologies 6 Comparison with Other Languages 7 Synopsis References Appendix 1 Additional Details on OCML 1 Functional Term Constructors SETORAL EAEE EEA ect act at AUN A RPA ac PING RIANA ee ALINE ret Ga edd Seat ET E I ANI YN Col Bape ee a ee So a le oR CEE A oO Le N TAT 3 Inheritance and Default Values 4 Interpreters and Proof System 4 1 THE OCML INTERPRETER FOR FUNCTIONAL TERMS ssccsessssesssescsseseeescsscsecseesseseceesesesscsesensess 4 2 THE OCMI PROOF SYST M sss 3 lt ssxdesties sh a aaa re a E aA E N EOE SEa RAEE e aE AES EEEE 4 2 1 Procedure for proving basic goal expressions in OCML sseesseeseeseneereeeeereren 4 2 2 Proof rules for non basic goal CXPVSSIONS scccscessecsecesecseceseeseceseeseceseeseceseeseeeaseneeeaeenge 35 35 36 40 41 44 43 INTRODUCTION This document is composed of three parts The first part describes WebOnto a tool providing web based visualisation browsing and editing support for developing and maintaining ontologies and knowledge models specified in OCML The description is user oriented in the sense that it is meant to provide guidance to a user rather than to describe the tool from a scholarly perspective The seco
68. s rel first 1 cons first 1 filter rest 1 rel filter rest 1 rel 5 These constraint checking mechanisms can be switched off if they are not required Knowledge Modelling in OCML Page 29 The above definition shows an example of the use of def functi on The OCML function filter takes as arguments a list and a unary relation re and returns the elements of which satisfy rel As illustrated by the definition the def option provides a declarative way of specifying a function the option body an effective way of computing its value for a given set of input arguments 2 3 OCML classes OCML also supports the specification of classes and instances and the inheritance of slots and values through isa hierarchies Classes are defined by means of a Lisp macro def cl ass which takes as arguments the name of the class a list possibly empty of superclasses optional documentation and a list of slot specifications as illustrated by the definitions in the next box These show a number of classes taken from the domain model for the Sisyphus I office allocation problem Linster 1994 This problem consists of allocating members of the YQT laboratory to the appropriate offices according to a number of constraints def class YQT member has project type project smoker type boolean cardinality 1 hacker type boolean cardinality 1 works with type YQT member belong
69. s to group type research group value yqt def class researcher YQT member def class secretary YQT member def class manager YQT member OCML provides support for the usual slot specification machinery which is found in frame based languages Specifically it provides the following slot options val ue A value which is inherited by all instances of a class default value A value which is inherited by all instances of a class unless overridden by other values type The value of this option should be a class say C This option specifies that all values of the associated slot should be instances of C It is also possible to specify that a slot value must belong to one or more classes say C1 Cn by using the notation or C1 C2 Cn max cardinality The maximum numbers of slot values allowed for a slot min cardinality The minimum numbers of slot values required for a slot cardinality The numbers of slot values required for a slot This option subsumes both min cardinality and max cardinality Knowledge Modelling in OCML Page 30 documentation The value of this option is a string providing documentation for a slot inheritance The inheritance mechanism used for dealing with default values If merge is used then all default values inherited from different ancestors are collected If supersede is used then default values inherited from more specific ancestors override
70. ssible output of the function or second argument of the binary relation no op iff def or and function f r forall args result gt apply f r args result holds relation result and binary relation f r forall x y gt holds f r x y holds relation y In the above definition it is worth highlighting the use of the special meta relation hol ds An expression such as holds lt r gt lt arg gt lt argy gt is satisfied if and only if the expression lt r gt lt arg gt lt argy gt is satisfied Thus holds has variable arity it can take one or more arguments In particular the number of additional arguments in a hol ds statement reflects the arity of the relation passed as first argument The relation hol ds has a special status because it is the only relation with variable arity supported by OCML 2 1 4 OCML relations summing up The set of relation options discussed here aims to provide a flexible and versatile range of modelling constructs supporting various styles of modelling While the emphasis is on operational modelling OCML also supports formal specification Moreover it provides facilities for integrating a specification with efficient proof mechanisms Ontolingua aware readers may have induced from the definition of the relation range that in contrast with Ontolingua OCML does not considers functions as relations This has to do with the operational nat
71. ssification mechanism which integrates a sophisticated concept definition language with rule based reasoning This approach allows a wider range of inferences to be drawn than those available from traditional frames trules systems such as KEE The existence of a powerful classifier is an important advantage that LOOM maintains over OCML On the other hand viewed purely as a knowledge modelling language LOOM exhibits a number of limitations When building knowledge models it may be necessary to make statements about the model being developed which do not have a direct inferential purpose For this reason OCML allows the inclusion of non operational statements and supports the specification of axioms about the current model In contrast LOOM only provides operational 9 The following points should not be construed as criticisms of the LOOM language This is primarily a knowledge representation language and should be of course judged with respect to knowledge representation criteria However given the high level of support provided by the language its organic relationship with the Ontolingua effort and its sound theoretical basis it makes sense to consider it as a plausible candidate for knowledge modelling Indeed LOOM has been used for ontological work Swartout et al 1996 Knowledge Modelling in OCML Page 44 constructs Another possible problem related to LOOM is that while it supports different knowledge representation parad
72. t cause any problem For instance if an if construct is encountered when expecting a functional term then an error will be generated if control terms are used in the body of the if The opposite however is not true functional terms can always be used in place of control terms Knowledge Modelling in OCML Page 25 Relation Option Role in Specification iff def Specifies both sufficient and necessary conditions for the relation to hold for a given set of arguments sufficient Specifies a sufficient condition for the relation to hold for a given set of arguments constraint Specifies an expression which follows from the definition of the relation and must be true for each instance of the relation def This is for compatibility with Ontolingua it specifies a constraint which is also meant to provide a partial definition of a relation axiom def A statement which mentions the relation to which it is associated It provides a mechanism to associate theory axioms with specific relations Table 2 1 Relation specification options in OCML 2 1 2 Operationally relevant relation options Relation options also play an operational role Specifically some relation options support constraint checking over relation instances while others provide proof mechanisms which can be used to find out whether or not a relation holds for some arguments Table 2 2 lists the relation options which are meaningful from an operational point of view and
73. ting from the body of a oop or repeat construct When a control term such as return term is encountered the most specific loop or repeat construct in the current execution stack is exited and the value obtained from evaluating term is returned return control term return term 3 INHERITANCE AND DEFAULT VALUES Generally speaking default values are values which apply unless other alternatives can be used In the OCML language the notion of default value is operationalized as follows Instances inherit values and default values from their superclasses down the inheritance hierarchy specified by i nstance of andsubclass of links Fora given slot says of a sample instance say the following scenarios can arise 1 1 has not inherited any default value In this case the value of s in I is given by all the values I has inherited from its superclasses plus any value locally specified for slot s of I 2 I has inherited some default values as well as non default ones In this case the default values are ignored and rule i is applied We say that the default values are overridden by the non default ones 3 I has inherited only default values and local values have been specified As in the previous case the default values are ignored and only the local values are considered 4 I has inherited only default values and no local values have been specified In this case there are two possibilities If the inheritance facet has not been s
74. ts In particular because these are relations it is possible to characterize them by means of the relation options discussed in section 2 1 1 For instance the definition below specifies the class of empty sets in terms of an iff def relation option def class EMPTY SET set set iff def not exists x element of x set 2 6 The generic Tell Ask interface 2 6 1 Tell a generic assertion making primitive OCML provides a generic assertion making primitive tell which provides a uniform mechanism for asserting facts independently of whether these refer to slot filling assertions new class instances or simply relation instances For example we can use tell to add a new value to the list of projects carried out by harry_c as follows 6 Formally a fact or assertion is a ground relation expression A relation expression is an expression such as lt r gt lt arg gt lt argn gt where lt r gt is a relation and arg is a functional term A ground expression or term is an expression or term which does not contain variables Knowledge Modelling in OCML Page 32 tell has project harry_c mt HAS PRO ECT HARRY_C MLT Analogously we can add a new instance of class researcher simply by stating tell researcher mickey m RESEARCHER MICKEY M 2 6 2 Ask a generic query posing primitive The relation centred view makes it possible to examine the contents of an OCML mode
75. ure of the language functions are dealt with by the OCML interpreter relations by the proof system 4 This constraint is only a limitation of the current implementation of the language in principle there is no reason why variable arity relations should not be supported Moreover in contrast with relations OCML functions can have variable arity To specify that a function can take an indefinite number of arguments OCML uses the same convention as Lisp the symbol amp r est is used before the last argument of a function argument schema Knowledge Modelling in OCML Page 28 2 2 OCML functions A function defines a mapping between a list of input arguments and its output argument Formally functions can be characterized as a special class of relations as in KIF Genesereth and Fikes 1992 However in operational terms there is a significant difference between a function and a relation functions are applied to ground terms to generate function values relation expressions can be asserted or queried Thus in accordance with the operational nature of OCML functions are distinguished from relations Functions are defined by means of a Lisp macro def function This takes as argument the name of a function its argument list an optional variable indicating the output argument as in Ontolingua this is preceded by an arrow gt optional documentation and zero or more function specification options These are def constraint body and lisp
76. ures Procedures define actions or sequences of actions which cannot be characterized as functions between input and output arguments For example the procedure below defines the sequence of actions needed to set the value of a slot These include a unassert statement which removes any existing value from the slot and a tell statement which adds the new value Both tell and unassert are procedures The former takes a ground logical expression and adds it to the current model The latter takes a relation expression and removes from the current model all assertions which match it Note that in accordance with the uniform view of a knowledge model slot changes are carried out by means of generic assertion and deletion operations i e in terms of tel andunassert def procedure SET SLOT VALUE i s v constraint and instance of i c slot of s c body do unassert list of s i any tell list of s i v 2 8 Rule based reasoning in OCML 2 8 1 Backward rules OCML also supports the specification of backward and forward rules A backward rule consists of a number of backward clauses each of which is defined according to the following syntax backward clause relation expression fi f logical expression Each backward clause specifies a different goal subgoal decomposition When carrying out a proof by means of a backward rule the OCML interpreter will try to prove the relevant goal by firing t
77. y defines relation has val ue but does not contribute to the specification def relation HAS VALUE p v dm Parameters have values w r t a particular design model iff def and parameter p desi gn model dm element of p v dm constraint or and exists vr has value range p vr element of v vr not exists vr has value range p vr prove by element of p v dm 2 1 3 A meta option for non operational specifications As shown above OCML provides a number of relation options which play both a specification and an operational role However in some cases we might want to use a Knowledge Modelling in OCML Page 27 keyword only for specification and not operationally for instance when we know that the value of the keyword in question is a non operational expression To cater for these situations OCML provides a special meta keyword no op which can be used to indicate that the enclosed relation option only plays a specification role An example of its use is shown by the definition of relation range which is shown below In the example the keyword no op is used to indicate that the iff def specification of the relation is not operational in particular it is not normally feasible to test the range of a function on all its possible inputs def relation RANGE f r relati on The range of a function or a binary relation is a relation which is true for any po

Knowledge Modelling in WebOnto and OCML A User Guide

Contents

Download Pdf Manuals

Related Search

Related Contents