A text-based model for opinion mining and sentiment analysis from online customer reviews in food industry

Abstract

In the rapid growth of technology and the Internet over recent years, e-commerce websites have been developed as a useful online media channel for users to easily make transactions such as online shopping and ordering food and drinks online, then share experience and feedbacks. Therefore, to be able to understand customer behaviors through positive or negative reviews about the products and services is an important desideratum. To offer a solution for this problem, the research proposes a method for customers opinion mining and sentiment analysis based on collecting data sets as customer reviews from the website Foody.vn – a top ranking website in the field of online ordering services. Machine learning models were conducted and evaluated to choose best model and then dashboards were created as visualizing results. The experimental results show that 90% accuracy of the proposed method; and valuable information and latent knowledge discovered from the corpus can support businessmen to capture the advantages and disadvantages of products and services and improve business with better strategies.

Keywords

• học máy
• khai phá ý kiến
• phân tích cảm xúc
• thương mại điện tử
• ý kiến khách hàng