Spam Comment Recognition of E-commerce Products Based on Collaborative Filtering Algorithm

In recent years, the way people express their opinions and communicate with each other has also changed. In the field of e-commerce product reviews, people increasingly prefer to post their opinions on products on shopping sites, and these user reviews contain a lot of useful information. At the same time, these views are full of useless and untrue spam, and the presence of such spam can affect the quality of product review mining. This paper aims to study a method for identifying spam reviews of e-commerce products based on collaborative filtering algorithm. This paper analyzes the definition of collaborative filtering algorithm, the advantages and disadvantages of collaborative filtering algorithm and the definition of spam in the field of product reviews, and also analyzes feature extraction methods and experimental dataset sources and obtained experimental results. The empirical analysis results show that there are various types of spam comments in e-commerce product reviews, and the existence of these spam comments affects people's important interests and affects the performance of automatic opinion mining systems. Therefore, analyzing and identifying spam comments is a very meaningful study. After adopting the collaborative filtering algorithm, the performance of the traditional e-commerce product spam comment identification method has been greatly improved, and we conducted a survey and found that more than 87% of people felt that the design of the e-commerce product spam comment identification method applied the collaborative filtering algorithm, and found that the design efficiency has become higher.