For doing business in this communication era, web is the best medium. For business owners and consumers, online businesses broke down the barrier of time and space compared to the physical shop or office. Big companies around the world are realizing that E-commerce (EC) in not just buying and selling over Internet rather improving the competence than other giants in the market. E-Commerce has allowed businesses to offer more choices to consumers. Increasing choice, however, has also increased the amount of data and information that consumers must process before they are able to select which items meet their needs. To grow their potential markets, the big shopping platforms like Amazon, Flipkart and ebay etc. wants to utilize Machine Learning (ML) potential to build unmatched competitiveness in the market. ML has empowered businesses to analyze all queries, whether searched or abandoned, from all the users. Application of machine learning for Predictive Analytics can enhance business opportunities by analyzing customer’s past click-through behaviour, purchases, preferences and history in real-time. To make fast real-time predictions from e-commerce data, the algorithm must be capable of processing huge volume of training data in reasonable time, and must be capable of handling large number of classes. So, the paper investigates the use of machine learning in E-commerce domain and its importance in predictive analysis. The need of Cloud platforms for analyzing E-commerce data is also established in this work. The paper concludes with exploration of potential areas of research in the field of E-commerce.