False Detection for Online Reviews by Using Deep Neural Network

Online reviews usually have the most significant impact on a customer's decision to buy a good or service. Therefore, manufacturers and retailers are anxious and concerned about customer feedback and reviews regarding this critical component and its impact on items. Unfortunately, the greater reliance on online reviews has an immediate effect on product sales since it may encourage false reviews that promote or disparage the goods and services. Opinion (Review) Spam happens when spammers fabricate and manipulate negative and false reviews to make money or gain an advantage. Machine learning approaches for categorization, and evolutionary algorithms for feature selection are used to extract meaningful information from text.Supervised learning techniques, which need labelled data, have been the focus of the most recent research. However, there needs to be more when addressing online review spam. Customer reviews on a product are becoming exponentially in number. The integrity of online opinions is a serious issue that is routinely disregarded. To make appropriate and accurate decisions or to gather marketing and commercial knowledge, it is crucial to have a technique for evaluating the trustworthiness of reviews. The product manufacturer gains knowledge about the benefits and drawbacks of their own and their competitors' goods, as well as consumer preferences and interests, which may help them decide how to maximise profits. Furthermore, it's crucial to find trustworthy consumer reviews. Bio-inspired algorithm-based feature selection is proposed to address the problem of categorising spam reviews.