York University at TREC 2008: Blog Track

Abstract

York University participated in the TREC 2008 Blog track, by introducing two opinion finding features. By initially focusing solely on the sentiment terms found in a document, using two different methods, and not considering the topic relevance, we saw an overall negative impact in the final evaluations. Post-TREC improvements show the necessity of combining opinion weights with the topic relevance scores. A tunable combination function is used, which revealed some interesting findings about the opinion finding features noting the strengths and weaknesses. We also introduce the Compass Information Retrieval system, which is based on Okapi, as the driver by which these experiments were conducted.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2008
Accession Number
ADA512713

Entities

People

  • Mladen Kovacevic
  • Xiangji Huang

Organizations

  • University of York

Tags

Communities of Interest

  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Automatic
  • Business Intelligence
  • Computer Science
  • Data Sets
  • Databases
  • Electronic Mail
  • Engineering
  • Frequency
  • Information Operations
  • Information Retrieval
  • Information Systems
  • Online Communications
  • Standards
  • Universities
  • Weighting Functions

Fields of Study

  • Computer science

Readers

  • Information Retrieval
  • Theoretical Analysis.

Technology Areas

  • AI & ML
  • AI & ML - Information Retrieval