Sentiment Analysis: Retrieving Positive Or Negative Polarities in News Articles Using Information Agents

Abstract

In this research, a sentiment analysis approach is proposed to extract sentiments associated with positive or negative polarities in news articles. Information retrieval agents were implemented to capture important keywords from the unstructured data such as news, distinguished the keywords into positive or negative lexicons, rank those keywords based on the frequency of occurrences, and recommend the news articles as positive or negative to the users. The proposed approach is illustrated with experimental results and their main implications are discussed.

Meeting Name

37th Annual Conference of Decision Sciences Institute

Department(s)

Business and Information Technology

Keywords and Phrases

Information Agent; Sentiment Analysis; Text Mining

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2006 Asia-Pacific Decision Sciences Institute, All rights reserved.

Publication Date

01 Jan 2006

This document is currently not available here.

Share

 
COinS