Abstract

Online reviews have become an important way for consumers to understand product quality and merchant services with the prosperity of online shopping. This paper adopts the method of natural language learning to conduct sentiment analysis on Amazon women's online shopping reviews, then establishes a bag of words model to vectorizes the text. Finally, a prediction model is established based on the Naive Bayes classifier, which predicts consumers' recommendation willingness by analyzing the review text. Through training and testing, the accuracy of the prediction model reaches 0.87, and improved to 0.92 by TF-IDF algorithm. This study analyzes the key influencing factors of women's online shopping which has a high reference value for merchants to improve services in a targeted manner to obtain more consumer recommendations.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call