A Study on Effective Product Marketing on E-Commerce based on Early Reviews

The mining of essential data from online sources has grown a famous study area in information retrieval methods in recent years. With the evolution and growth of social media, it is an excellent inclination for people to experience what they are exploring with friends across multiple social networking stages. A considerable number of favorable product reviews are available online and each minute, a massive amount of unstructured data is instantly updated. The unstructured data is in the form of text which collected from conferences, reviews, and social media. The information is valuable for new users to determine whether the assistance satisfy their requirements before sharing product reviews. Reviews in online have become a crucial cause of information for users before creating an acquainted investment judgment. The product purchase depends on the early reviews. In this paper, Enhanced Margin-Based Embedding Ranking Model (EMERM) is proposed for forecasting of early reviewer. Various parameters such as order behaviour, correlation of their review with product demand etc., are used to assess the characteristics of early reviewers. The experiments are conducted on two various e-commerce Amazon and yelp datasets. The experimental result shown that the proposed approach outperforms several competitive baselines.