An Optimal Aggregation of Product Data using Vector Space Model
Susmitha Gunti1, Krishnamoorthi Makkithaya2, Deepthi S.3
1Susmitha Gunti, Department of Computer Science and Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
2Krishnamoorthi Makkithaya, Professor, Department of Computer Science and Engineering, Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India.
3Deepthi S, Assistant Professor, Department of Computer Science and Engineering at Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, Karnataka, India
Manuscript received on January 14, 2020. | Revised Manuscript received on January 28, 2020. | Manuscript published on February 10, 2020. | PP: 2003-2007 | Volume-9 Issue-4, February 2020. | Retrieval Number: D1409029420/2020©BEIESP | DOI: 10.35940/ijitee.D1409.029420
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: In the current scenario there exists many versions of a particular product. A product might be laptop, mobile or any other gadget. With increase in number of versions there is a need to analyze the reason for release of the new version of the product. This can be done by the study of reviews and ratings provided by consumers. To get a more accurate output we first relate the rating and review using Sentiment Analysis (SA). SA is a form of text mining that helps us to understand the attitude and behavior of a customer towards a product/service. The ratings given by the customer may not be in the same level of agreement as in the review text. Customer may have issues with the product and has explained in the review but can be generous and give decent rating, such circumstances often depends on the emotional quotient of the customer. Therefore, there is a need for a system which can elicit the polarity among the reviews and check if there is proper agreement between the ratings and reviews given by user till the product become obsolete. In order to provide the correlation between the ratings and reviews lexicon method of sentiment analysis is used to generate the sentiment score for each review. Based on the sentiment score obtained the reviews are further classified into extreme negative, negative, neutral, positive, and extreme positive and compared to the ratings given by the customer. With this reviews as input, feature selection is done using vector space model. The output obtained depicts the success factors and failures of a product which helps to build a better version.
Keywords: Sentiment Analysis, Text Mining, Lexicon Approach, Feature Selection, Vector Space Model
Scope of the Article: Smart Spaces