Journal of Shanghai University(Natural Science Edition) ›› 2014, Vol. 20 ›› Issue (3): 289-295.doi: 10.3969/j.issn.1007-2861.2014.02.013

• no • Previous Articles     Next Articles

Co-training Algorithm for Quality Analysis of Online Customer Reviews

JIN Jian1,2, JI Ping2,3   

  1. 1. School of Government, Beijing Normal University, Beijing 100875, China;
    2. Shenzhen Research Institute, The Hong Kong Polytechnic University, Shenzhen 518057, Guangdong, China;
    3. Department of Industrial & Systems Engineering, The Hong Kong Polytechnic University, Hong Kong 999077, China
  • Received:2014-05-03 Online:2014-06-26 Published:2014-06-26
  • Contact: 季平(1961—), 男, 副教授, 博士, 研究方向为企业资源规划、运营管理、优化理论及其应用. E-mail: p.ji@polyu.edu.hk E-mail:p.ji@polyu.edu.hk

Abstract: Online customer reviews exist widely on e-commerce websites. Customer concerns and preferences about products are involved in these reviews. It is critical for business professionals to analyze online reviews efficiently and effectively in the fierce market competition. Typically, the quality analysis of online reviews is a good example. Accordingly, two aspects of features are identified from online reviews, and a Co-training algorithm is built to analyze quality of online reviews. Effectiveness of the algorithm and its advantages over a single classification/regression algorithm is confirmed by experiments.

Key words: Co-training algorithm, online customer review, product design, quality of online reviews, text mining, data quality

CLC Number: