International Journal of Engineering Research in Electronics and Communication Engineering

Product Recommendation from Textual Reviews Using Hybrid Filtering Approach

Author : Sunil Jadhav ¹ Ritambhara Rajeshirke ²

Date of Publication :20th April 2018

Abstract: The amount of information on the internet grows rapidly and people need some system to find and access appropriate information. Recommender Systems (RS) are currently used in both the research and in the commercial areas. RS are information filtering systems that deal with the problem of Information Overload. The existing recommendation is based on POI (Point of Interest), Geographical location, User Preference learning and algorithms like LDA, OGRPL (Online Graph Regularized User Preference learning) are used for information extraction and also use the Attribute Pruning (AP), Frank-Wolfe algorithm for improving the performance of the system with some limitation like high retraining cost, unable to capture change in preferences, work for specific value of k. The proposed system recommendation is base on sentiment prediction from textual review using the LDA for feature extraction and is originally based on Hybrid Filtering Approach. K numbers of products are recommended to the user. Hence, the proposed system improves the prediction accuracy of the recommendation system

Reference :

1. Xiaojiang Lei, Xueming Qian, Member, and Guoshuai Zhao,”Rating Prediction based on Social Sentiment from Textual Reviews”,IEEE TRANS,2016
2. Zhou Zhao, Deng Cai, Xiaofei He and Yueting Zhuang,”User Preference Learning for Online Social Recommendation”,DOI 10.1109/ TKDE. 2016.2569096, IEEE Trans., 2016.
3. Hongzhi Yin, Xiaofang Zhou, Bin Cui, Hao Wang, Kai Zheng and Quoc Viet Hung Nguyen,” Adapting to User Interest Drift for POIRecommendation”,IEEE Trans,2016.
4. Xavier Amatriain, Alejandro Jaimes, Nuria Oliver, and Josep M. Pujol,” Data Mining Methods for Recommender Systems”, ICREA,Chapter 2,page no.41-49
5. Yining Teng, Lanshan Zhang, Ye Tian, Xiang Li,” A Novel FAHP Based Book Recommendation Method by Fusing Apriori Rule Mining”, International Conference, 2015
6. Kishor Barman and Onkar Dabeer,” Analysis of a Collaborative Filter Based on Popularity Amongst Neighbors,” IEEE Transactions on information theory, vol. 58, no. 12, december 2012.
7. Jianxun Liu, Mingdong Tang, Member,Zibin Zheng, Member, Xiaoqing (Frank) Liu, Member, Saixia Lyu,”Location-Aware and Personalized Collaborative Filtering for Web Service Recommendation”, IEEE Transactions ,2015.
8. G. Linden, B. Smith, and J. York,” Amazon.com Recommendations: Item-to-Item Collaborative Filtering,” IEEE Internet Comput., vol. 7, no. 1, pp. 76–80, 2003.
9. Alastair A. Abbott and Ian Watson,” OntologyAided Product Classification: A Nearest Neighbour Approach”, Department of Computer Science, University of Auckland,Auckland, New Zealand
10. Kang Liu, Liheng Xu, and Jun Zhao,” CoExtracting Opinion Targets and Opinion Words from Online Reviews Based on the Word Alignment Model”, IEEE TRANS, VOL. 27, NO. 3, MARCH 2015.