Predictive Analysis of E-Commerce Products

For the past few years, there has been increasing trend for people to buy products online through e-commerce sites. With the user-friendly platform, there is loop hole which does not guarantee satisfaction of the customers. The customers have the habit of reading the reviews given by other customers in order to choose the right product. Due to high number of reviews with mixture of good and bad reviews, it is confusing and time-consuming to determine the quality of the product. Through these reviews, the vendors would also want to know the future trend of the product. In this paper, a predictive analysis scheme is implemented to detect the hidden sentiments in customer reviews of the particular product from e-commerce site in real-time basis. This serves as a feedback to draw inferences about the quality of the product with the help of various graphs and charts generated by the scheme. Later, an opinion will be drawn about the product on the basis of the polarity exhibited by the reviews. Finally, prediction over the success or failure of the product in the regular interval of the timestamp is done using time series forecasting method. A case study for iPhone 5s is also presented in this paper highlighting the results of rating generation, sentiment classification, and rating prediction.