Feature Based Sentiment Analysis of Product Reviews using Modified PMI-IR method

  IJCTT-book-cover
 
International Journal of Computer Trends and Technology (IJCTT)          
 
© 2016 by IJCTT Journal
Volume-34 Number-2
Year of Publication : 2016
Authors : Sanjay Kalamdhad, Shivendra Dubey, Mukesh Dixit
  10.14445/22312803/IJCTT-V34P120

MLA

Sanjay Kalamdhad, Shivendra Dubey, Mukesh Dixit "Feature Based Sentiment Analysis of Product Reviews using Modified PMI-IR method". International Journal of Computer Trends and Technology (IJCTT) V34(2):115-121, April 2016. ISSN:2231-2803. www.ijcttjournal.org. Published by Seventh Sense Research Group.

Abstract -
In online product reviews users discuss about products and its features. A product may have hundreds or thousands of reviews, consumers share their experience about products and comments about products characteristics. These product reviews may have positive or negative sentiments. A positive sentiment contains good opinion about product and its features similarly a negative sentiment tells drawbacks and problems of product and its features. Feature may be part of the product or its characteristics. In this paper we use modified PMIIR method for analyzing the sentiments in online product reviews about the various features of products. We download the product reviews from internet using the web crawler and stored it in inverted index format. Using the parts-of speech tagging, extract the two-word opinion phrases and calculates the semantic orientation by measuring the mutual information between each phrases and positivity and negativity. Summary of sentiments of each feature is presented based on average semantic orientation value. Summarization shows the sentiment classification of features of products.

References
[1] Turney, P. 2002 Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised Classification Reviews ACL‟02.
[2] Hatzivassiloglou, V and McKeown, K 1997. Predicting the Semantic Orientation of Adjectives, In proc of 35th ACL/8th EACL.
[3] Q. Ye, Y. Li, Z. Yiewn, 2005. Semantic Oriented Sentiment Classification For Chinese Products Reviews: An Experimental Study on Books and Cell Phone Reviews.
[4] Z. Zi-qiong, LI Yi-jun, YE Qiang, LAW Rob, International Conference 2008, Sentiment Classification for Chinese Product Review Using an Unsupervised Internet-based Method.
[5] X. DUAN, T. HE, Le SONG, Research on Sentiment Classification of Blog Based on PMI-IR.
[6] Turney P. Mining the Web for Synonyms: PMI-IR versus LSA on TOEFL.
[7] M. Hu and B. Liu, Mining and summarizing customer reviews, KDD‟04 2004.
[8] X. Ding, B. Liu and PS. Yu, 2008 International conference, A Holistic Lexicon Based Approach to Opinion Mining. [9] Vapnik V. N. The Nature of Statistical Learning Theory, New York: Springer 1998.
[10] Fei, Z. C., Liu J., Wu G.F., Sentiment Classification using Phrase Patterns. In: Proceeding of the 4th International Conference on Computer and Information Technology (CIT‟04). Wuhan, China: IEEE, 2004: 1-6.
[11] Maite Taboada‟s SO-CAL program, Lexicon-based methods for sentiment analysis M Taboada, J Brooke, M Tofiloski, K Voll [12] NLProcessor - Text analysis toolkits 2000. https://www.infogistics.com/textanalysis.html
[13] Won Young Kim, Joon Suk Ryu, Kyu Il Kim, Ung Mo Kim, A Method for Opinion Mining of Product Reviews using Association Rules.
[14] Santorini, B. 1995. Part-of-Speech Tagging guideline for the Penn Treebank Project (3rd revision, 2nd Printing), Technical Report, Department of Computer and Information Science, University of Pennsylvania.
[15] A.-M. Popescu, O. Etzioni. Extracting product features and opinions from reviews[C]//Proc. of Conf. on Empirical Methods in Natural Language Processing, EMNLP‟05, 2005: 339-346.
[16] M. Gamon, A. Aue. Automatic identification of sentiment vocabulary: Exploiting low association with known sentiment terms[C]//Proc. of the ACL-05 Workshop on Feature Engineering for Machine Learning in Natural Language Processing, 2005:57- 64.
[17] Church. K. W. and Hanks, P. 1990, Word Association Norms, Mutual Information and Lexicography.
[18] T. Mullen, N. Collier. Incorporating topic information into sentiment analysis models [C]//Proc. of the ACL 2004 on Interactive poster and demonstration sessions, 2004.
[19] Turney and Littman 2003, Measuring praise and criticism: Inference of semantic orientation from association.
[20] V. Ng, S. Dasgupta and S. M. Niaz Arifin, Examining the Role of Linguistic Knowledge Source in Automatic Identification and Classification of Reviews. ACL‟06, 2006.
[21] S. Kim and E. Hovy Determine the Sentiment of Opinions.
[22] B. Pang, L. Lee, and S. Vaithyanathan Thumbs up? Sentiment Classification Using Machine Learning Techniques EMNLP‟2002.

Keywords
Sentiment analysis, web crawler, semantic orientation, PMI-IR, summarization.