Identification of Fake Reviews: A Review

In recent years, artificial intelligence (AI) approaches have witnessed rapid progress in the context of computers and smartphones, which has led to a number of exciting new applications. In many cases, these methods make it possible for the gadgets to carry out their functions in an intelligent manner. Utilizing common AI techniques such as machine learning, deep learning methods, natural language processing, knowledge representation as well as expert systems, may help the targeted apps become more intelligent and efficient. As the practice of shopping online becomes increasingly popular, an increasing number of people are considering making their essential purchases from companies that operate exclusively online. This method of shopping does not demand a significant amount of the customer's time. A customer enters an online store, conducts a search for the product that meets his or her needs, and then makes a purchase. Customers have difficulty acquiring items from Internet retailers owing to the low quality of the products offered in these stores. Customers will not make orders for a product unless they have first looked at its rating and read any reviews that have been posted about it. Reviews can be genuine or fake and can be either negative or positive. The opinions of other people are a source of happiness for the buyer who purchased the new items. In this case, a single negative fake review may influence the customer's decision whether to purchase the goods or not. At the same time, positive fake reviews give misleading information about the product and may lead to customers purchasing a low-quality product. In most cases, it is conceivable that this one review is a forgery. So, we have to remove this type of fake reviews and provide the users with the original reviews and ratings related to the products. Many methods are proposed by researchers around the globe. This paper reviews a few of such ideas proposed.