Review-Based Personalized Recommendation System in Emergency Management

In emergency management, it is important to distribute supplies to people that meet their needs and interests. A good personalized recommendation system must rely on users’ real interests. Relied on purchase history, overall ratings and other forms of data, which are far from enough to infer users’ real interests, the limitation of traditional recommendation methods are revealed. In this paper, we proposed a review-based personalized recommendation system which could be applied in various areas, especially in emergency management. This method extracts a user’s latent interest and preference from his/her reviews of a product. The system aims to mine the features of the product which the users paid the most attention to, then find user group that shares similar interest, finally recommend the products that can most satisfy the users’ (or decision makers’) needs. An experiment was conducted and the result demonstrated that our system could generate a reliable and realistic recommendation.

[1]  Gerhard Friedrich,et al.  Recommender Systems - An Introduction , 2010 .

[2]  Lillian Lee,et al.  Opinion Mining and Sentiment Analysis , 2008, Found. Trends Inf. Retr..

[3]  Gediminas Adomavicius,et al.  New Recommendation Techniques for Multicriteria Rating Systems , 2007, IEEE Intelligent Systems.

[4]  Bing Liu,et al.  Mining and summarizing customer reviews , 2004, KDD.

[5]  Francesco Ricci,et al.  Improving recommender systems with adaptive conversational strategies , 2009, HT '09.

[6]  Loriene Roy,et al.  Content-based book recommending using learning for text categorization , 1999, DL '00.

[7]  Bing Liu,et al.  Web Data Mining: Exploring Hyperlinks, Contents, and Usage Data , 2006, Data-Centric Systems and Applications.

[8]  Tiejian Luo,et al.  Towards an Introduction to Collaborative Filtering , 2009, 2009 International Conference on Computational Science and Engineering.

[9]  Zhang Jing.  Personalized recommendation algorithm based on product features , 2007 .

[10]  Yong Zhang,et al.  An incident information management framework based on data integration, data mining, and multi-criteria decision making , 2011, Decis. Support Syst..

[11]  Yue Lu,et al.  Latent aspect rating analysis on review text data: a rating regression approach , 2010, KDD.

[12]  Sung-Shun Weng,et al.  Feature-based recommendations for one-to-one marketing , 2003, Expert Systems with Applications.

[13]  Jonathan L. Herlocker,et al.  Evaluating collaborative filtering recommender systems , 2004, TOIS.

[14]  Olfa Nasraoui,et al.  Web Usage Mining , 2011 .

[15]  David M. Pennock,et al.  Mining the peanut gallery: opinion extraction and semantic classification of product reviews , 2003, WWW '03.

[16]  Gediminas Adomavicius,et al.  Toward the next generation of recommender systems: a survey of the state-of-the-art and possible extensions , 2005, IEEE Transactions on Knowledge and Data Engineering.

[17]  Bing Liu,et al.  Opinion observer: analyzing and comparing opinions on the Web , 2005, WWW '05.