Sentiment Analysis on E-Commerce Product Reviews Using Machine Learning and Deep Learning Algorithms: A Bibliometric Analysis, Systematic Literature Review, Challenges and Future Works

Abstract

The success of every company is based on the service provided to customers, so it is essential to know the perception of buyers in relation to specific products, this allows the company to analyze concrete data that can be invaluable for its development in a highly competitive market, and improve the quality of products. The purpose of this study is to gain a deeper understanding about the most used items, methods to test the performance of the model, techniques and those that had better accuracy, metrics and programming languages used in the sentiment analysis on e-commerce products review. An exhaustive search was carried out in 4 databases: ScienceDirect, Scopus, ProQuest and Web of Science using keywords, identifying 20 articles between 2018 and 2024. This article was based on the PRISMA methodology, taking into account the inclusion and exclusion criteria. To then make a synthesis of the findings related to: Items of application of sentiment analysis, methods to test the performance of the model, techniques, metrics and programming languages; and propose new challenges and future work. Products of different items was the most item used, Cross validation the most used method to test model, also the most efficient techniques were Support Vector Machine (SVM) and LSTM had better accuracy, the essential metrics were F1-Score, and finally the most used programming language to develop the models was Python. In this study, we offer a comprehensive view of machine learning and deep learning algorithms to analyze sentiment in e-commerce product reviews, so we identify existing challenges and suggest future directions to improve the efficiency of these methods, thus contributing to providing more effective and accessible solutions in the field of sentiment analysis, We also highlight how this analysis can serve as a source of competitive advantage by providing useful information for companies to improve the quality of their products.

Keywords

• Sentiment analysis
• E-Commerce
• Product
• Machine learning
• Deep learning