Sentiment Analysis using machine learningalgorithms: online women clothing reviews
Shuangyin Xie · NORMA · 2019
Internet technology has been closely related to life. It not only convenient people's lives but also allows people to share information, especially in the field of e-commence. People leave message and share their feelings online. As a result, sentiment analysis becomes more and more attracted. Accurate sentiment analysis not only allows customers to better understand the product, but also enables the company to get better feedback from the market. In this paper, we use data set from online women clothing reviews to conduct sentiment analysis, which can be downloaded from Kaggle. The machine learning methods used in this research are Support Vector Machine, Logistic Regression, Random Forest, Naive Bayes. All experiments were done in this research using python. We evaluate the model in terms of accuracy, precision, recall, F1-score and Area Under Curve(AUC). This study provides us with sentimental analysis of various women clothing opinions dividing them Positive, Negative and Neutral behaviour. These data suggest that the Naive Bayes gives highest accuracy to classify the Reviews, which is 93%. Keywords: sentiment analysis, machine learning, Support Vector Machine, Logistic Regression, Random Forest, Naive Bayes.