Today’s digital era is characterized by enormous amounts of online information, which keeps accumulating on a daily basis. The sheer size of the user-made content has created a complicated issue for businesses which is not falseness but its spread. The emerging result is that, with so much information, the impact of falsehood increases greatly. Thus, most of the data that businesses have to mine are just unreliable. In order to mine ultimate truthfulness from unreliable data, effective mining techniques must be applied. This paper presents a system based on artificial intelligence using a Naïve Bayes Algorithm with the purpose of detecting false product reviews. The system’s principal objective is to differentiate true and untrue reviews in order to extract only useful information from opinion mining procedures. Unlike conventional text summarization, however, our approach is for the collection of consumer opinions on products based on which we will present a comparison of products and the best consumer sentiment, positive or negative.