Volume 9 Number 7 (Jul. 2014)
Home > Archive > 2014 > Volume 9 Number 7 (Jul. 2014) >
JSW 2014 Vol.9(7): 1952-1959 ISSN: 1796-217X
doi: 10.4304/jsw.9.7.1952-1959

Study on Identification of Subjective Sentences in Product Reviews Based on Weekly Supervised Topic Model

Wei Jiang

College of Computer Science and Technology, Wuhan University of Technology, Wuhan, China; Department of Information Engineering, City College of Wuhan University of Science and Technology, Wuhan, China

Abstract—Sentiment analysis or opinion mining in online product reviews is a method that can automatically detect subjective information regarding the entity such as opinions, attitudes, and feelings expressed by consumers. Online product reviews always include objective and subjective sentences; identification of subjective sentences in the given content is a very important and foundational task in the research of opinion mining. In this paper, we focus on the problem of identification of sentence-level subjective sentences, propose a weakly supervised model mixed topics based on LDA for identification of the subjective sentences, considering the impact of multiple topic factors on the identification of subjective sentences. The approach exploits semi-supervised learning method, and extended the existing basic LDA topic model for the identification of subjectivity in text. This work iterates the model prior probability by using a small domain-independent lexicon. Finally, the proposed model is applied to a online review corpus and the experimental shows that the proposed model can effectively improve the recognition effect.

Index Terms—online product reviews, opinion mining, topic model, subjective sentence


Cite: Wei Jiang, "Study on Identification of Subjective Sentences in Product Reviews Based on Weekly Supervised Topic Model," Journal of Software vol. 9, no. 7, pp. 1952-1959, 2014.

General Information

ISSN: 1796-217X (Online)
Frequency:  Quarterly
Editor-in-Chief: Prof. Antanas Verikas
Executive Editor: Ms. Yoyo Y. Zhou
Abstracting/ Indexing: DBLP, EBSCO, CNKIGoogle Scholar, ProQuest, INSPEC(IET), ULRICH's Periodicals Directory, WorldCat, etc
E-mail: jsweditorialoffice@gmail.com
  • Mar 01, 2024 News!

    Vol 19, No 1 has been published with online version    [Click]

  • Jan 04, 2024 News!

    JSW will adopt Article-by-Article Work Flow

  • Apr 01, 2024 News!

    Vol 14, No 4- Vol 14, No 12 has been indexed by IET-(Inspec)     [Click]

  • Apr 01, 2024 News!

    Papers published in JSW Vol 18, No 1- Vol 18, No 6 have been indexed by DBLP   [Click]

  • Nov 02, 2023 News!

    Vol 18, No 4 has been published with online version   [Click]