Opinion Retrieval Project

Collaborative Research
Handling Negations and Temporal Aspects for Opinion Retrieval

This project is sponsored by research grants from the National Science Foundation .

Principal Investigator at University at Illinois at Chicago

PI: Prof. Clement Yu
Department of Computer Science
University at Illinois at Chicago
NSF grant number: IIS-0842546

Principal Investigator at SUNY at Binghamton

PI: Prof. Weiyi Meng
Department of Computer Science
State University of New York at Binghamton
NSF grant number: IIS-0842608

Project Overview

Opinions can have major impact on our society. They can affect the sales of products, the change of government policy, and even people's vote in elections. Thus, it is of high significance to study opinion retrieval and analysis. In the age of the Web, more and more people choose to express their opinions on a wide range of topics on the Web in the forms of blogs, product/service reviews, and comments. Opinion retrieval aims to retrieve or mine opinions about different topics from text documents.

In this project, two important problems in opinion retrieval will be investigated. The first problem is to determine whether and when a negation word flips the sentiment of an opinionated word. The second problem is to examine how the opinions about a topic change over time and what causes significant changes. Both problems are critical for accurate opinion retrieval. Failure to correctly determine the impact of negation words can lead to incorrect conclusions while failing to discover the shift of sentiment on important topics may have major consequences in national security and personal lives. The interactions between these two issues are also to be studied in this project. These problems will be studied in three different domains - consumer products and services, medical/health care and politics.

We believe that the ability to precisely identify the reasons behind significant sentiment shifts in a timely manner, together with more accurate interpretation of sentiments as a result of better handling of negation words, will make opinion retrieval much more practically useful, which in turn will make opinion retrieval a much more attractive research area not only for computer scientists but also for political scientists, social scientists and linguists. Thus, this project has the potential to lead to transformative changes in opinion retrieval research.

Any opinions, findings and conclusions or recomendations expressed on this sites are those of the PIs and do not necessarily reflect the views of the National Science Foundation (NSF).

Other Participants

The following students have participated/are participating in this project:

Related Publications

  1. Yiyao Lu, Weiyi Meng, Wanjing Zhang, King-Lup Liu, and Clement Yu. Automatic Extraction of Publication Time from News Search Results. 2nd International Workshop on Challenges in Web Information Retrieval and Integration (WIRI2006), pp.141-150, Atlanta, Georgia, April 2006.
  2. Wei Zhou, Clement Yu, Neil Smalheiser, Vetle Torvik and Jie Hong. Knowledge Intensive Conceptual Retrieval and Passage Extraction of Biomedical Literature. ACM SIGIR Conference, pp.655-662, July 2007.
  3. Wei Zhang, Clement Yu, Weiyi Meng. Opinion Retrieval from Blogs. ACM Sixteenth Conference on Information and Knowledge Management (CIKM), pp.831-840, Lisboa, Portugal, November 2007.
  4. Wei Zhang, Lifeng Jia, Clement Yu, Weiyi Meng. Improve the Effectiveness of the Opinion Retrieval and Opinion Polarity Classification. ACM 17th Conference on Information and Knowledge Management (CIKM 2008), poster paper, pp.1415-1416, Napa Valley, California, October 2008.
  5. Lifeng Jia, Clement Yu, Weiyi Meng. The Effect of Negation on Sentiment Analysis and Retrieval Effectiveness. 18th ACM Conference on Informationand Knowledge Management (CIKM 2009), Hong Kong, China, pp.1827-1830, November 2009.

Last change: November 17, 2009 / meng@cs.binghamton.edu