Mining and Summarizing Customer Reviews

Authors- Minqing Hu and Bing Liu

Year – 2004

Published in- Proceedings of the 10th ACM International conference on knowledge discovery and data mining. 

Link - http://sifaka.cs.uiuc.edu/course/591cxz04f/peng1.pdf

Importance to my Research – Very High

MY REVIEW

This paper proposes a product review classification system that downloads reviews from the web and classifies them as positive and negative based on each product feature. In doing so, the framework automatically identifies different features about a product or service being reviewed and then mines the relevant sentence to look for opinions. Finally the system presents the user with a summary of reviews for one particular product. This paper downloaded the product reviews from Amazon and C|Net for their experiments. The proposed system can be used by product manufactuers for improving their products as well as customers in deciding which product to buy.

Before I begin the review of this paper I would say this is one of the best written papers, it is so well organized and thought about before writing, that any questions that I had in mind were answered at the right time. I should congratulate the authors for their excellent effort in writing this paper, which made my reading a pleasurable experience.

Some Future Directions

1. It would be a good idea to provide some additional statistics based on the review classification e.g. I would be interested in knowing how many people say that the quality of picture is acceptable under bright sunlight but not very good in dim light. I guess the current system only identified picture quality as a feature and puts all the reviews related to the picture quality underneath. Am I correct?

more details coming soon…

Cite this article as
Critical Review on “Mining and Summarizing Customer Reviews” by V. Potdar, 10th Mar, 2008. Available Online – http://drvidy.wordpress.com/2008/03/10/mining-and-summarizing-customer-reviews/

(more…)