Title :
Chinese Review Spam Classification Using Machine Learning Method
Author_Institution :
Sch. of Manage., Tianjin Univ., Tianjin, China
Abstract :
With great development of the e-commerce, the number of product reviews grows rapidly on the e-commerce website. Review mining has recently received a lot of attention, which aims to discover valuable information from the massive product reviews. An important subject of review mining is review spam classification, which classifies reviews into reviews or spam reviews, offering high-quality data to review mining. In this paper, we first present a categorization of Chinese review spam, and then classify the reviews by using machine learning method with different features, finally analyze the impact of different features on classification performance. The experiments show that Chinese review spam classification will obtain high accuracy by using machine learning method with appropriate features.
Keywords :
Web sites; data mining; electronic commerce; learning (artificial intelligence); pattern classification; unsolicited e-mail; Chinese review spam categorization; Chinese review spam classification; e-commerce Website; machine learning method; product reviews; review mining; Accuracy; Classification algorithms; Data mining; Feature extraction; Mobile handsets; Semantics; Support vector machines; Chinese review spam classification; Review mining; feature Identification; machine learning;
Conference_Titel :
Control Engineering and Communication Technology (ICCECT), 2012 International Conference on
Conference_Location :
Liaoning
Print_ISBN :
978-1-4673-4499-9
DOI :
10.1109/ICCECT.2012.200