Abstract

With the growing popularity of review platforms, the phenomenon of "review bombing", consisting of issuing negative ratings in a short period of time, has become a significant problem, affecting the reputation of products and services. The paper presents a literature review on review bombing detection techniques using natural language processing (NLP) and machine learning tools. Various approaches and models that allow for the identification of undesirable rating patterns in text data are discussed. The analyzed technologies include classification methods, anomaly detection and sentiment analysis techniques, as well as their effectiveness in the context of different types of data. The paper points out current challenges related to review bombing detection, such as language diversity, rating contextuality and evolving fraud techniques. Finally, directions for future research are presented that can help improve the effectiveness of existing solutions and develop new, innovative detection methods.

Download

Share

COinS