Biblio

With the development of the Internet, people are interested to share their views and opinions about the product on the web, forums, blogs etc. These online reviews are important for individual users and organization. Recently, it is a common tendency to the user to read the reviews or comments before purchasing some products or services. The online reviews are helpful for the business organizations in order to promote their product. However, in practice, these online reviews may be fake in order to promote or devalue the product. These fake reviews are called as opinion spam. Objective of the research paper is that, to select the best feature subset for detecting the fake review. To select a small subset of features out of the thousands of feature is important for accurate classification of review spam detection. Therefore, a good feature selection method is needed in order to speed up the processing rate, predictive accuracy. In this paper hybrid improved Binary Particle Swarm optimization (iBPSO) and cuckoo search (CS) is used for feature selection and Naive Bayes and k Nearest Neighbor classifier is used for classifying the review as spam and ham. Experimental results have shown that the proposed algorithm has yielded the best performance compared with the swarm intelligence techniques Binary Particle Swarm Optimization (BPSO) and Shuffled Frog Leaping (SFL).