1 / 14

Dan Su Department of Computer Science Brigham Young University

Towards Semantic Web: An Attribute-Driven Algorithm to Identifying an Ontology Associated with a Given Web Page. Dan Su Department of Computer Science Brigham Young University. Motivation. Semantic Web Enrich current web manually will be laborious, tedious and error-prone Ontology

Download Presentation

Dan Su Department of Computer Science Brigham Young University

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. Towards Semantic Web: An Attribute-Driven Algorithm to Identifying an Ontology Associated with a Given Web Page Dan Su Department of Computer Science Brigham Young University

  2. Motivation • Semantic Web Enrich current web manually will be laborious, tedious and error-prone • Ontology Capture the semantics of information from various sources and output a concise description The number of different ontologies is increasing

  3. given a web page, how could you identify which ontology in the ontology library will be associated with it?

  4. Ontology Library • It is necessary to classify ontology in an ontology library system in order to facilitate searching, managing and re-using ontology

  5. Text Categorization • Assign a Boolean value to each pair <dj, ci> Є D * C • Traditional approaches • Advantage of traditional approaches • Disadvantage of traditional approaches

  6. Why not a direct matching? • Direct Matching • Reasons for discarding a direct matching: 1. The increasing large number of ontologies 2. Ignore weights of different attributes

  7. Thesis Statement • focus on identifying the ontology associated with the given web page from an ontology library based on the attributes similarity computation • discuss the feasibility of an ontology-based machine learning approach

  8. Overview of Method

  9. Assumption of Ontology Library • Unique identification: a unique URI or a unique name • Life-time: the valid period for current ontology version • Domain concept and the associated attributes

  10. Automatic Construction of Training corpus • <html> • <p> Honda MH-2 1994</p> ……… </html> • ----------------------------------- • <html> • <ontology car.ontology> </ontology> • <p><attri make> Honda </attri> <attri model>MH-2</attri><attri year> 1994</attri> • </html>

  11. Attributes Similarity Computation Attribute Recognizer <html> <p> Honda price $9000, model…</p> ……… </html> make

  12. TFIDF-like computation Oj= (wj1, wj2, ………wjn) wk=afjk * idfjk

  13. Evaluation • Precision and Recall • Compare our results with the results using Pure TFIDF algorithm.

  14. Contribution • Automatically identify the ontology associated with a give web page • Advance the transformation from current web to a semantic web • Can be extended to the text categorization field

More Related