Opinion-Based imprecise query answering

  • Authors:
  • Muhammad Abulaish;Tanvir Ahmad; Jahiruddin;Mohammad Najmud Doja

  • Affiliations:
  • Department of Computer Science, New Delhi, India;Department of Computer Engineering, New Delhi, India;Department of Computer Science, New Delhi, India;Department of Computer Engineering, New Delhi, India

  • Venue:
  • PAKDD'10 Proceedings of the 14th Pacific-Asia conference on Advances in Knowledge Discovery and Data Mining - Volume Part II
  • Year:
  • 2010

Quantified Score

Hi-index 0.00

Visualization

Abstract

There is an exponential growth in user-generated contents in the form of customer reviews on the Web But, most of the contents are stored in either unstructured or semi-structured format due to which distillation of knowledge from this huge repository is a challenging task In addition, on analysis we found that most of the users use fuzzy terms instead of crisp terms to express opinions on product features Considering these facts, in this paper, we present an opinion-based query answering framework which mines product features and opinionated words to handle user queries over review documents The proposed framework uses BK-FIRM (Bandler-Kohout Fuzzy Information Retrieval Model) that facilitates the formulation of imprecise queries using linguistic qualifiers, retrieves relevant opinion documents, and presents them in the order of their degree of relevance The efficacy of the system is established through experiments over customer reviews on different models of digital camera, and mp3 player.