Personalized Recommendation Algorithm Based on Product Reviews

Personalized Recommendation Algorithm Based on Product Reviews

Zhibo Wang (International School of Software, Wuhan University, Wuhan, China), Mengyuan Wan (International School of Software, Wuhan University, Wuhan, China), Xiaohui Cui (International School of Software, Wuhan University, Wuhan, China), Lin Liu (School of Software, East China University of Technology, Shanghai, China), Zixin Liu (School of Software, East China University of Technology, Shanghai, China), Wei Xu (School of Software, East China University of Technology, Shanghai, China) and Linlin He (School of Software, East China University of Technology, Shanghai, China)
Copyright: © 2018 |Pages: 17
DOI: 10.4018/JECO.2018070103

Abstract

Under the background of leap-forward development for the internet, e-commerce has played an important role in people's daily life, but huge data sizes have also brought problems, such as information overload which can be solved by using a recommendation system effectively. However, with the development of the e-commerce, the amount of the product catalogs and users becomes larger, which causes lower performance of the traditional recommendation system. This article comes up with a personalized recommendation algorithm based on the data mining of product reviews to optimize the performance of the new recommendation system. Features of the product were extracted, for which the users' sentiment polarity was analyzed. This article develops a recommendation system based on the user's preference model and the product features to get the recommendation result. Experimental results show that a personalized recommendation has significantly improved the accuracy and recall rate when compared with a traditional recommendation algorithm.
Article Preview
Top

Relate Work

In personalized recommendation algorithm, there are three more prevailing, they are based on the content, collaborative filtering and combinations. Content-based recommendation algorithm is mainly faced with a lack of some abstract feature extraction capacity, cold start and was unable to discover new points of interest and other issues. Collaborative filtering can discover new points of interest, but also to deal with some non-regular items, but collaborative filtering is also facing the problem of cold start, sparsity, identity issues and scalability issues.

Using a combination of recommendation algorithm can effectively avoid the impact of some of the defects of single algorithm recommendation result is too large. Recommended combination algorithm has been extensively studied. Rojsattarat and Soonthornphisaj (2003) resolved based on the content and composition recommendation algorithm collaborative filtering basic framework, and use the recommended algorithm based on the contents of SVM optimized combination of properties recommendation algorithm. Liu, Lai and Lee (2009) proposed hybrid algorithm based on association rules and collaborative filtering can solve simple use of collaborative filtering algorithm cannot demand changes for users to make rapid and effective response of the drawbacks of this method based KNN-CF's strategy to block the user to purchase goods sequential data prediction algorithm applied to improve the performance of the recommendation system. Of course, apart from the several recommendation algorithms, there are many other recommendation algorithms in practice there are more widely used. Kim and Kim (2003) proposed a recommendation algorithm based on the model, and to optimize the use of multi-level association rules recommendation algorithms are often faced with sparse data and to get higher than collaborative filtering algorithm performance. Desrosiers and Karypis (2011) put forward a recommendation algorithm based on the nearest neighbor, and put it with the collaborative filtering algorithm was compared, by way of dimensionality reduction and data relationships graph-based analytical methods to optimize the performance of the system.

Since the study recommendation algorithm in principle has been more mature, so the recent research on the recommendation algorithm is mainly optimized for various types of user recommendation system problems made, Liu et al. (2008) using the optimized similarity calculation model eliminate the impact of overheating of the mainstream commodity recommendation system, making collaborative filtering recommendation algorithm accuracy rate of up to 25%. Rodríguez et al. (2010) based collaborative filtering algorithm incomplete preference relations model to solve the recommendation system cold start problems, and achieved certain results.

Complete Article List

Search this Journal:
Reset
Open Access Articles
Volume 18: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 17: 4 Issues (2019)
Volume 16: 4 Issues (2018)
Volume 15: 4 Issues (2017)
Volume 14: 4 Issues (2016)
Volume 13: 4 Issues (2015)
Volume 12: 4 Issues (2014)
Volume 11: 4 Issues (2013)
Volume 10: 4 Issues (2012)
Volume 9: 4 Issues (2011)
Volume 8: 4 Issues (2010)
Volume 7: 4 Issues (2009)
Volume 6: 4 Issues (2008)
Volume 5: 4 Issues (2007)
Volume 4: 4 Issues (2006)
Volume 3: 4 Issues (2005)
Volume 2: 4 Issues (2004)
Volume 1: 4 Issues (2003)
View Complete Journal Contents Listing