Abstract: In recent years, the border between Online Media and E-Commerce is diminishing. Almost every single person
in a metropolitan daily uses both social media like Face-book, Twitter, etc. for networking and uses internet to make
huge purchases using e-commerce sites like Flipkart, Amazon, etc. We often login to e-commerce web-sites using our
social accounts like FB or G+. We can also share our recent purchase details on the social media using the links to the
product pages of e-commerce sites. We are focusing on the product recommendation to the users on e-commerce sites by
leveraging the information or knowledge gained from the users social accounts. This will enable to assess the needs of the
user in cold start situations. Cold Start is a state when user logs in to the e-commerce website for the first time and we
don’t have any information about the history of purchases, shopping trends, etc. as it is not yet created or available.
When we have users social account information (no con-fidential information will be accessed) like posts, friends, shares,
etc. then we can harness this to our benefit. For example, we will be applying data mining algorithms to access the microblogs
the user has created and extract the useful keywords and hence this data from the micro blogs becomes the basis
for product recommendation in cold start situations.