1 / 18

Ontology learning: state of the art and open issues

Ontology learning: state of the art and open issues. Lina Zhou Presenter : Taizhi Li. Overview. L earning-oriented model of ontology development F ramework for ontology learning Ontology learning approaches Classification schema of domains Open issues Conclusion. Question s ? ? ?.

tyler
Download Presentation

Ontology learning: state of the art and open issues

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Ontologylearning:stateoftheartandopenissues LinaZhou Presenter:TaizhiLi

  2. Overview • Learning-oriented model of ontology development • Framework for ontology learning • Ontology learning approaches • Classification schema of domains • Open issues • Conclusion

  3. Questions??? • WhatisOntologyLearning? • OntologyLearningreferstotheautomaticdiscoveryandcreationofontologicalknowledgeusing machine learning techniques. • Whatisadvantagesofontologylearningcomparedwithcrafting ontologies? • Alargerscaleandafasterpace. • Mitigatehuman-introducedbiasesandinconsistence.

  4. Major issues in ontology development • Ontology representation: adequacy and inference efficiency • Ontology acquisition (incomplete,subjective,outdated) eg.Dictionaries,webdocuments,databaseschema • Ontology evaluation(content,technology,methodology,application) completeness,consistency,correctness;learning • Ontology maintenance(organize,search,updateexistingontologies) eg.SWOOGLE

  5. Learning-oriented model RODvs.ConventionalOntologyDevelopment? Iterativeprocess Rapid ontology development (ROD)

  6. RapidOntologydevelopment

  7. Requirementsforintegratedtools Functions that can facilitate ontology development: • Knowledge elicitation • Ontology retrieval • Ontology editing • Ontology validation • Collaborative development • Ontology transformation and presentation

  8. Framework for ontology learning

  9. Ontology organizationsteps B A A+B • Clusteringsynonymoustermsandtheirrelations • Derivinginverserelations • Discoveringlocalcentersofconcepts • Buildinghigher-levelontologies C C  B A c* p j i

  10. Ontology learning approaches

  11. Taxonomyoflearningtechniques

  12. Classificationschemaofdomains Domains Poorquality Learningstrategies Approachesselection: • Established vs. under-developeddomains Top-downBottom-up Knowledge-richKnowledge-lean (eg.biology)(eg.communitydevelopment)

  13. Classificationschemaofdomains • Emergingvs. conventionaldomains Bottom-upTop-down Hybrid • Technology-heavyvs.technology-lightdomains (Hybrid) • Self-containedvs.interdisciplinarydomains (Top-level)

  14. Open issues • Human understandable vs. machine-understandable • Learning specific relations eg.Part-wholeforSequenceOntologyforgenomicannotation; Related_synonymforGeneOntology • Learning higher-degree relations eg.Trustrelation:whotrustswhomonwhat • Learning definitions eg.isdefinedas;isreferredtoas

  15. Open issues • Term filtering Suggestion: mutual information and traditional term weighting tech; contrast analysis; anaphora and co-reference resolution • Mapping to high-level ontology Suggestion: path analysis; treat the top-level ontologies as seeds • Evaluation benchmark Suggestion: benchmark corpora • Incremental ontology learning Suggestion: update existing ontologies incrementally

  16. Open issues • Levels of ontology learning eg. “Dell Notebook”canbeasubclassoraninstanceof“Notebook” • Multi-agent learning Suggestion:agent-basedparadigm • Learning beyond text Suggestion:content-based imageretrievaltechs

  17. Conclusion • Ontology operational within web and distributed system • Semanticweb • Ontologies:turnthecurrentwebsiteintoanetworkofknowledgeresourcesandservices

  18. Thankyou!

More Related