• DocumentCode
    260546
  • Title

    Analysis of retrieval result on ontology-based query reformulation

  • Author

    Azizan, Azilawati ; Abu Bakar, Zainab ; Noah, Shahrul Azman

  • Author_Institution
    Fac. of Comput. & Math. Sci., Univ. Teknol. MARA, Shah Alam, Malaysia
  • fYear
    2014
  • fDate
    2-4 Sept. 2014
  • Firstpage
    244
  • Lastpage
    248
  • Abstract
    Users normally facing difficulties to expresses their information needs into a query format that the search system can use to process. Therefore users frequently modify their query with intention to retrieve better results. The goal of this experiment is to compare the precision of the retrieved results by employing domain ontology to reformulate the user query. We proposed to reformulate the query (initially in natural language) by replacing it with the terms retrieved from ontology. Then we improve the approach by combining the terms from ontology with keyword/s extracted from the initial user´s natural language query. Precision measure had been conducted for six queries and the results showed that the idea of combining ontology terms with keyword from query gave a promising result. It achieved higher precision in comparison when using ontology terms alone. This experimental result indicates that query reformulation is more efficient when ontology term/s combined with key element from the user´s query.
  • Keywords
    ontologies (artificial intelligence); query processing; search engines; domain ontology; key elements; keyword extraction; ontology-based user query reformulation; precision measure; query format; query ontology terms; retrieval result analysis; search system; user natural language query; Agriculture; Conferences; Natural languages; Ontologies; Search engines; Semantics; ontology; precision measure; query reformulation;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Computer, Communications, and Control Technology (I4CT), 2014 International Conference on
  • Conference_Location
    Langkawi
  • Print_ISBN
    978-1-4799-4556-6
  • Type

    conf

  • DOI
    10.1109/I4CT.2014.6914183
  • Filename
    6914183