Text Summarization is an active and interesting research area that has emerged to in recent times .There is an increase in trend of online shopping. Before buying any product or service online users prefer to read opinions about that product of service. But problem is that there are millions of reviews for same product is available over different websites and users do not have the time to read all the reviews. So need of review summarization is there. Text summarization method summarizes the content of reviews of people with help of similarity and clustering methods and guide them whether to purchase that product or not. Text summarization can be of many types. This research work proposes an extractive improved text summarization method which performs better than existing text summarization methods. This is done by including some improvements in the text summarization method like in this research work, those sentences are retained which have at least one noun, one adjective, one verb and one adverb. This stops the elimination of some of the important reviews. Also this research work is combining STASIS similarity and LDA technique for calculating semantic and context similarity respectively. After that similarity score generated by both the techniques is combined and an overall similarity score is calculated. Sentences are assessed using this similarity score and conflicting sentences are eliminated. And most important improvement of this research work is to improve k-means clustering by including Levenshtein distance instead of Euclidean distance. After doing this improvements both the existing and improved text summarization methods are applied on datasets of reviews and their performance is compared using factors like Rand Measure, Precision, Recall, F-measure, Review Importance Factor and it is proved that proposed method is better than existing method.

Text Summarization, Data Mining,Extractive Summarization,Text Mining

[1] Abhishek Kauirik, Sudhanshu Naithani, “A comprehensive study of text mining approach”, IJCSNS International Journal of Computer Science and Network Security, Vol. 16,Isuue 2,Feb 2016
[2] Alaa F. Alsaqer, Sreela Sasi. “Movie review summarization and sentiment analysis using rapidminer”, 2017 International Conference on Networks & Advances in Computational Technologies, ISSN: 978-1-5090-6590-5, July 2017
[3] D. Gaikwad and C. Mahender, "A review paper on text summarization", International Journal of Advanced Research in Computer and Communication Engineering, Vol. 5,Issue 3, March 2016
[4] A. Shetty, R. Bajaj, "Auto text summarization with categorization and sentiment analysis", International Journal of Computer Applications,Vol.130, Issue 7 ,2015
[5]Xiaoye Wanga, Xiaorui Chaia, Ching-Hsien Hsua, Yingyuan Xiaoa, Yukun Lia,"cluster analysis based on opinion mining", 8th International Conference on Ubi-Media Computing,ISSN:978-1-4673-8270-0,2015
[6]https://onlinecourses.science.psu.edu/stat555/node/86
[7] A.Dharmarajan, T.Velmurugan, “Applications of partition based clustering algorithms: a survey" , IEEE International Conference on Computational Intelligence and Computing Research, ISSN:978-1-4799-1597-2,2013
[8] Preeti Arora, Dr. Deepali, Shipra Varshney,"Analysis of k-means and k-medoids algorithm for big data", International Conference on Information Security & Privacy, Dec 2015
[9]S. Sethi, A. Dixit,“ An adaptive web search system based on web usages mining” International journal of computer engineering and application , Vol. X , Issue. 1, ISSN: 23213469 ,2016
[10] S. Sethi, A. Dixit, “An Automatic User Interest Mining Technique for Retrieving Quality Data” International Journal of Business Analytics. Vol. 4, Issue. 2, pp 62-79, ISSN: 2334-4547, 2017