ì î SIG-SWO-A201-02 OWL ƒp Semantic Web
Ý Ý ÝÛ Ú Û ÌÍÍÛ ì Web 90-: ñå Tom Gruber ~ (Ontolingua) ì (KIF) Generic Ontology CYC, WordNet, EDR PSM Task Ontology 95-97: XML as arbitrary structures 97-98: RDF 98-99: RDFS (schema) as a frame-like system 00-01: DAML+OIL 02-07: OWL OWL On-To-Knowledge OIL ~ EU OIL ² DAML OIL ~ DAML+OIL»~ DAML+OIL W C DAML+OIL WebOnt Working Group û OWL
RDF S) OWL RDF(S) DAML+OIL OWL class-def subclass-of class-expressions slot-def AND, OR, NOT subslot-of slot-constraints domain has-value, value-type range cardinality slot-properties trans, symm DAML Ontology Library (Ontology's by Keyword) http://www.daml.org/ontologies/keyword.html Keyword URI academic department http://www.cs.umd.edu/projects/plus/daml/onts/cs1.0.daml academic department http://www.cs.umd.edu/projects/plus/daml/onts/cs1.1.daml Academic Positions http://www.daml.ri.cmu.edu/ont/homework/cmu-ri-employmenttypesont.daml access primitives control http://www.w3.org/2000/10/swap/pim/doc.rdf acronym http://orlando.drc.com/daml/ontology/thesaurus/call/current/ activity http://www.kestrel.edu/daml/2000/12/operation.daml Actors http://opencyc.sourceforge.net/daml/cyc.daml Actors http://www.cyc.com/2002/04/08/cyc.daml Actors http://www.cyc.com/cyc-2-1/cyc-vocab.daml address book http://www.w3.org/2000/10/swap/pim/contact.rdf agenda http://www.daml.org/2001/10/agenda/agenda-ont
Baseball Ontology 7/1 # 2 38 995/,2 47 -,80-, -,80-, 439 2 38 /,2 995/,2 47 /,2 Г4 2 38 4 0/ 995 2.82,3,.: 4 4 0/ 2 38 7/1 995 477/18 39, 38 2 38 7/18 995 47 7/18. 02, /,2 394 4 7/1,-4:9 7/18.422039-,80-, 7/18.422039 7/18.422039-,80-, 4394 4 7/18.422039 /,2 ;078 43 314 /,2 ;078 43 314 /,2 394 4 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439!489$0,843,2 7/18,-0!489$0,843,20 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,904 0/.70,9 43,90 7/18 8:-,88 1 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439,20 7/18,88 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439 33 3 7/18,-0 33 3 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,90 4 0/.70,9 43,90 7/18 8:-,88 1 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439 70,90;039 7/18,88 7/18 8:-,88 1 7/18 8:-,88 1 /,2 #0897.9 43 /,2 43!745079 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 4393:2-07 /,2 43!745079 /,2 94,88 7/18,88 7/1,-4:9 995 47 %# 2 8. 02,897 3 7/18,88 /,2 94,88 /,2 #0897.9 43 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439 0, :0 7/18,-0 0, :0 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,90 4 0/.70,9 43,90 7/18 8:-,88 1 /,2 #0897.9 43 /,2 43!745079 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 439/ ; 8 43 /,2 43!745079 /,2 94,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439 ; 8 43 7/18,88 /,2 94,88 /,2 #0897.9 43 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439!07843 7/18,-0!07843 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,904 0/.70,9 43,90 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439!, 07 7/18,-0!, 07 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,904 0/.70,9 43,90 7/18 8:-,88 1 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 43925 4 00 7/18,88 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439#0 01 7/18,-0 #0 01 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,90 4 0/.70,9 43,90 7/18 8:-,88 1 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439 30:5;039 7/18,88 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439$0,843 7/18,-0 $0,843 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,904 0/.70,9 43,90 7/18 8:-,88 1 /,2 #0897.9 43 /,2 43!745079 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 439 0,7 /,2 43!745079 /,2 94,88 7/18,88 7/1,-4:9 995 47 %# 2 8. 02,897 3 7/18,88 /,2 94,88 /,2 #0897.9 43 7/18 8:-,88 1 7/18,88 7/18,88 7/1,-4:9 995/,2 47 -,80-, -,80-, 439$07 08 7/18,-0 $07 08 7/18,-0 7/18.422039 7/18.422039 4 0/.70,9 43,904 0/.70,9 43,90 7/18 8:-,88 1 /,2 #0897.9 43 /,2 43!745079 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 439 420 /,2 43!745079 /,2 94,88 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 439%0,2 /,2 #0897.9 43 7/18 8:-,88 1 7/18 8:-,88 1 /,2 #0897.9 43 /,2 43!745079 7/1 7084:7.0 995/,2 47 -,80-, -,80-, 439,, /,2 43!745079 /,2 94,88 å å(
Issue Real software processes change dynamically... very difficult to maintain need the facility to correspond to changing process dynamically First, Approach develop software process ontologies through the domain analysis using two case studies Afterwards, design the system which generate software processes interactively with a user Goal A Process-Centered Software Engineering Environment Using Ontologies Development of Software Process Ontologies Step1. Step2. Step3. Extract real software processes and objects from the two cases Organize extracted processes and objects into distinct groups such that there are more similarity in meaning, while defining the software process scheme A input Process reference B output A C A+B A or + B Invent Join Append Extract Give detailed definitions to each processes using software process scheme
Process Ontology activity Invent Join Append Extract Focus on input,output, and reference roles Invent with references Invent without references Join with references Join without references Evaluate Review Examine Extract with references Extract without references Focus on reference, tool and agent roles Software Process Ontologies(conceptual hierarchy) Process Ontology including about 250 concepts Object Ontology including about 400 concepts
Start Object Software Process Automation using ontologies Goal Object Retrieve processes satisfied with constraints Process Ontology Object Ontology
Case Study A query about S P What SEP between the following input and output? Input : Requirement Specification from Output : Detailed Design of Each Component of Initial Software Process Plan Candidates Output Specify Testing S/W Integration Detailed Design of Each Input Requirement Specification from Design Basic Design Basic I/F Subsystems Review All Basic Designs Together Design Detailed Design Detailed I/F Subsystems Review All Detailed Designs Together Evaluate All Detailed Designs Specify Testing Units of S/W Design Logical Aspects of DB Evaluate All Basic Designs Design Physical Aspects of DB
Interim Software Process Plan Candidates Output Specify Testing S/W Integration Detailed Design of Each Input Requirement Specification from Design Basic Design Basic I/F Subsystems Design Detailed Design Detailed I/F Subsystems Review All Detailed Designs Together Evaluate All Detailed Designs Specify Testing Units of S/W Design Logical Aspects of DB Evaluate All Basic Designs Design Physical Aspects of DB Final Software Process Plan Candidates Output Specify Testing S/W Integration Detailed Design of Each Input Requirement Specification from Design Basic Design Basic I/F Subsystems Design Detailed Design Detailed I/F Subsystems Review All Detailed Designs Together Evaluate All Detailed Designs Specify Testing Units of S/W Design Logical Aspects of DB Evaluate All Basic Designs Design Physical Aspects of DB
Evaluation Applicability It turned out that the environment generated SPP candidates good for the user s query with user interaction Issues and Future work The environment can t allocate human and time resource The environment has not the facility that supports the user to add user s specific processes The methodology for building ontologies has no theoretical aspects,such as soundness g å(
How to Build up Domain Ontologies with Less Cost Taking Existing Information Resources DODDLE Project (a Domain Ontology rapid DeveLopment Environment) exploiting a MRD Constructing up just a conceptual hierarchy Why not taking MRD to build up domain ontologies? Good News: MRD have many concepts. Bad News: The concepts have been defined from the point of natural language processing (common sense) and so there are some concept drift between MRD and specific domain ontologies.
3 + 2 Concept Drift 3 + 2 "/97+, / +68 " 3 + 2 36/97+, /4+68,/-+97/30-32-/48.6 08 DODDLE '36./8 Domain Expert +8- /. "/79 872+ 7 7 Trimmed- Results Analysis Extension of Modification
Matched-Results Analysis Root 3:/ 3:/ Best-Match Internal Node #8+ 3:/ Trimmed-Results Analysis A B C $6 /.4+68 D A 0 0 3 B C D $6 /. 3./ %7/66/ -327869-87 8 /79, 86// A B C D 2 8 + 3./
Acquiring not only Conceptual Hierarchy but Concept Definitions DODDLE II Project (Extending DODDLE) Info. Resource:Domain-Specific Texts Technique: Co-occurence information WordSpace
WordNet DODDLE-II Overview Extracting Concept Relationships from Domain-Specific Texts WordSpace Marti A. Hearst, Hinrich Schutze Words and phrases in texts can be expressed by vector representation containing co-occurrence statistics Inner products among the vectors work as the similarity between the words and phrases. high value with similar words coming up around the concepts
Constructing WordSpace WordSpace Texts (4-gram array) Texts (4-gram array) Texts (4-gram vector array).. f8 f4 f3 f7 f8.. f8 f4 f3 f7 f8.. f8 f4 f3 f7 f8 f4 f1 f3 f8 f9 f2 f5 f1 f7 f1 f5.. f4 f1 f3 f4 f9 f2 f5 f1 f7 f1 f5.. f4 f1 f3 f4 w f2 f5 f1 f7 f1 f5.. w w w1 w3 C w2 w4 WordNet synset Collocation Matrix Vector representation for each 4-gram Context vector w A sum of 4-gram vectors around w Word Vector W A sum of w in Texts Vector representation of a concept C A sum of W in same synset Constructing Concept Specification Template non-taxonomy? TAXONOMY non-taxonomy? non-taxonomy?
DODDLE-II alpha on Perl/Tk (UNIX platform) IIalpha The Target Domain A Case Study Contracts for the International Sale of of Goods(CISG) Input Terms to TR module 46 Legal Terms from CISG Part-II Domain-Specific Texts to NonTR module full text of CISG (about 10,000 words)
Results 46Terms Select Best Matches Trimming 377Terms 113Terms Modification 56Terms 61Terms The paths from Spell Matched Nodes to the Root of WordNet Initial Model Trimmed Model Legal Ontology Taxonomic Relationships for input terms constructed from TR Acquisition Module
Support Ratio Support ratio: How match is included in final domain ontology the intermediate products at each DODDLE activity. Extracting Concept Relationships setting value for WordSpace Extraction frequency of 4-gram (times) 8 Collocation scope (4-grams) 10 Context scope (4-grams) 60 the concept pairs extracted according to context similarity (threshold 0.9993)
Domain Experts Modifying Concept Specification Templates ex) non-taxonomic Relationships for assent non-taxonomy? : offeror TAXONOMY : act non-taxonomy? : effect non-taxonomy? : offer non-taxonomy? : person non-taxonomy? : offeree non-taxonomy? : withdrawal non-taxonomy? : time TAXONOMY : proposal AGENT LEGAL-SEQUENCE ANTONYM : person : offer : withdrawal Recall & Precision of Concept Specification Templates recall precision The number of extracted concept pairs
Conclusions DODDLE II: A Domain Ontology Construction Support Environment using MRD and Domain-Specific Texts Legal experts appreciate DODDLE II to some extent. Future Work How to Decide Threshold for CS How to Identify NTR Another DM method instead of WordSpace Another Case Studies with Large Scale OWLå OWLå UMLå OWL RDF(S) RDF»