Learning Knowledge Rich User Models from the Semantic Web
- 1. Learning Knowledge Rich User Models from the Semantic Web Gunnar Aastrand Grimnes First Year Talk 14th May, 2003
- 2. Presentation Overview Motivation Preliminary Experiments Agentcities & GraniteNights The Future
- 3. Motivation The Semantic Web should: Facilitate learning from the Web. Facilitate reuse of learning outcomes. Hypothesis : Learning from data annotated with semantic mark-up should outperform learning from traditional (HTML) Web. Goals: The learned model should be expressed in a Semantic Web Language. Such a learned model should be re-usable across domains and applications.
- 4. Preliminary Experiments Compare performance of learning from plain text and from semantic meta-data. Using traditional ML algorithms as baseline approach: Naïve Bayes K-Nearest Neighbour Explore application of more knowledge intensive approaches, such as ILP (Progol). An Empirical Investigation of Learning From the Semantic Web, Pete Edwards, Gunnar AA. Grimnes and Alun Preece – Presented at Semantic Web Mining Workshop at ECML/PKDD, Helsinki, 2002
- 5. Issues Datasets in a Semantic Web language were very hard to come by. We used two datasets: ITTalks (Seminars described using HTML vs. DAML+OIL). Citeseer (Full text of Academic Papers vs. BibTex converted to RDF). How does RDF map to an instance representation suitable for learning?
- 6. Results Largely negative. K Nearest Neighbour on plain-text had best accuracy. … but: 10 lines of RDF vs. 6000 words of full-text paper. Reasons for failure: Shallow and artificial RDF. Statistical methods used. Progol results were the most interesting: % Classifying Machine Learning papers: inClass(A) :- publisher(A,'Morgan Kaufmann'), booktitleword(A,learning).
- 7. Agentcities & the Evening Scenario EU funded – 5th F.W. In Aberdeen since January’02. WeatherAgent online since February’02. Evening Scenario City Nodes Tourist Information Recommendations The fun has just started: OpenNET
- 8. GraniteNights Raison d’être: Agentcities Agent Technology Competition. Need a Semantic Web framework for learning user profiles. Bring together different people/research areas in the department: agents, learning, scheduling, constraints, etc. Proof that RDF is usable! GraniteNights - A Multi-Agent Visit Scheduler Utilising Semantic Web Technology, Gunnar AA. Grimnes, Stuart Chalmers, Pete Edwards and Alun Preece Submitted to CIA2003
- 11. Query By Example RDQL too complicated to write by hand. Query by example is very intuitive. Internal conversion to RDQL. Could be “smarter” than RDQL. <q:Query> SELECT ?x WHERE (?x, ?y, ?z), <q:template> <akt:Academic> ( ?x, <rdf # type>, <akt # Academic> ), <akt:family-name> ( ?x, <akt # family-name>, "Brown" ) Brown </akt:family-name> </akt:Academic> </q:template> </q:Query>
- 12. QbEx with constraints <q:Query> <q:template> <r:Restaurant> <r:type rdf:resource=“r#Tandoori" /> <r:open-time> <cif:Variable rdf:ID="x"> <cif:varname>x</cif:varname> </cif:Variable> </r:open-time> </r:Restaurant> </q:template> <q:constraints> <cif:Comparison> <cif:comparisonOperator>></cif:comparisonOperator> <cif:comparisonOp1> <cif:Variable rdf:about="#x"/> </cif:comparisonOp1> <cif:comparisonOp2> <cif:Integerconst> <cif:constantValue>1900</cif:constantValue> .. . .
- 13. GraniteNights Profiling <ep:User rdf:about=“profileagent#gunnar” ep:name=“gunnar” ep:pword=“****”> <ep:preference> <q:Query> <q:template> <pub:EnglishPub> <pub:servesBeer rdf:resource=“#flowers”/> </pub:EnglishPub> ... <ep:interactions> <rdf:Seq><rdf:li> <ep:Interaction ep:timestamp=“20030508T135013”> <ep:pref> <q:Query> <q:template> <pub:EnglishPub> <pub:servesBeer rdf:resource=“#flowers”/> </pub:EnglishPub> ... <pub:EnglishPub> <pub:servesBeer rdf:resource=“#hobgoblin”/> ... <pub:EnglishPub> <pub:servesBeer rdf:resource=“#flowers”/> ...
- 14. GraniteNights Profiling II Current implementation: Most frequently specified constraint. Possible improvements: Super/Sub-class inference in the ontology, i.e. Flowers and Hobgoblin are both sub- classes of Real Ale. Combination of constraints important, i.e.Pete likes Lager when eating Curry, but Ale for his occasional pub-visit. Requires more sophisticated techniques than counting.
- 15. The Future User modelling in a broader scope: User roles, commitments etc. Learning from RDF: Generalisation. Case-based reasoning. RDF as model language. Learning Knowledge Rich User Models from the Semantic Web, Gunnar AA. Grimnes To appear in Doctoral Consortium, User Modeling 2003, Pittsburgh, July 2003.
- 16. Questions ?