Spam reviews are increasingly appearing on the Internet to promote sales or defame competitors by misleading consumers with deceptive opinions. This paper proposes a co-training approach called CoSpa (Co-training for Spam review identification) to identify spam reviews by two views: one is the lexical terms derived from the textual content of the reviews and the other is the PCFG (Probabilistic Context-Free Grammars) rules derived from a deep syntax analysis of the reviews. Using SVM (Support Vector Machine) as the base classifier, we develop two strategies, CoSpa-C and CoSpa-U, embedded within the CoSpa approach. The CoSpa-C strategy selects unlabeled reviews classified with the largest confidence to augment the training dataset to retrain the classifier. The CoSpa-U strategy randomly selects unlabeled reviews with a uniform distribution of confidence. Experiments on the spam dataset and the deception dataset demonstrate that both the proposed CoSpa algorithms outperform the traditional SVM with lexical terms and PCFG rules in spam review identification. Moreover, the CoSpa-U strategy outperforms the CoSpa-C strategy when we use the absolute value of decision function of SVM as the confidence.

Abstract:With the rapid development of electronic commerce, spam reviews are rapidly growing on the Internet to manipulate online customers' opinions on goods being sold. This paper proposes a novel approach, called CoFea (Co-training by Features), to identify spam reviews, based on entropy and the co-training algorithm. After sorting all lexical terms of reviews by entropy, we produce two views on the reviews by dividing the lexical terms into two subsets. One subset contains odd-numbered terms and the other contains even-numbered terms. Using SVM (support vector machine) as the base classifier, we further propose two strategies, CoFea-T and CoFea-S, embedded with the CoFea approach. The CoFea-T strategy uses all terms in the subsets for spam review identification by SVM. The CoFea-S strategy uses a predefined number of terms with small entropy for spam review identification by SVM. The experiment results show that the CoFea-T strategy produces better accuracy than the CoFea-S strategy, while the CoFea-S strategy saves more computing time than the CoFea-T strategy with acceptable accuracy in spam review identification.