Phani K. Cheruku, Atul Kumar
Abstract: In this paper we present an empirical study of fake reviews detection algorithm.With the increase in internet usage, the demand for online servicing is growing rapidly, this leads to some threats like fake review. The users who used a product or service may give a genuine review, which makes it useful when other customers search for product/services. Whereas the online fake review may damage the customer sentiments and leads to negative impact on the product or services. Users’ opinions are the main source of reviews for selected products or services. To get profit or popularity for a services or brands fake reviews are generally written to advance or downgrade the targeted items. Existing systems studied fake reviews but a strong detection technique is needed in this problem. The service sectors like restaurants, e-commerce product selling websites have significant impact on their business through reviews, their customers increase when the reviews are good and vice versa. This proposed system examines detecting fake reviews that have been evaluated in the Yelp restaurant domain.
Keywords: Support Vector Machine, Term Frequency, Logistic Regression, Natural language processing, K-Nearest Neighbor, Decision Trees