一种上下文感知的E-commerce评级大数据赋权方法

A context-aware weighting approach for big data of quality ratings in E-commerce

电子商务(E-commerce)的飞速发展,产生了大量针对商品的在线评级数据,通过分析评级数据,用户可以对商品的质量进行评估.然而,评级数据的海量性和差异性使得用户难以快速而准确地评估商品的质量.鉴于此,提出一种基于E-commerce评级的上下文感知赋权方法(context-aware weighting approach,CWA),以选出少数"重要"的评级数据并抛弃大多数"不重要"的评级数据,从而确保商品质量评估的快速性和准确性.最后,通过一组实验验证了CWA的有效性.

With the fast development of E-commerce, large amounts of quality rating data for commodities are generated online. By analyzing the rating data, users can evaluate the commodities＇ quality. However, due to the massiveness and diversity of the rating data, it is a challenge for users to evaluate the commodity quality quickly and accurately. To this end, a context-aware weighting approach for E-commerce ratings, context-aware weighting approach （CWA） is proposed. With CWA, a few important rating data are selected and most unimportant data dropped. Thus the commodity quality can be evaluated quickly and accurately. A series of experiments validate effectiveness of the proposed CWA.