IFSpard: An Information Fusion-based Framework for Spam Review Detection

Online reviews, which contain the quality information and user experience about products, always affect the consumption decisions of customers. Unfortunately, quite a number of spammers attempt to mislead consumers by writing fake reviews for some intents. Existing methods for detecting spam reviews mainly focus on constructing discriminative features, which heavily depend on experts and may miss some complex but effective features. Recently, some models attempt to learn the latent representations of reviews, users, and items. However, the learned embeddings usually lack interpretability. Moreover, most of existing methods are based on single classification model while ignoring the complementarity of different classification models. To solve these problems, we propose IFSpard, a novel information fusion-based framework that aims at exploring and exploiting useful information from various aspects for spam review detection. First, we design a graph-based feature extraction method and an interaction-mining-based feature crossing method to automatically extract basic and complex features with consideration of different sources of data. Then, we propose a mutual-information-based feature selection and representation learning method to remove the irrelevant and redundant information contained in the automatically constructed features. Finally, we devise an adaptive ensemble model to make use of the information of constructed features and the abilities of different classifiers for spam review detection. Experimental results on several public datasets show that the proposed model performs better than state-of-the-art methods.