The Web is a vast pool of resources which comprises of a lot of web pages covering all aspects of life. Understanding a user’s interests is one of the major research areas towards understanding the web today. Identifying the relevance of the surfed web pages for the user is a tedious job. Many systems and approaches have been proposed in literature, to try and get information about the user’s interests by user profiling. This paper proposes an improvement in determining the relevance of the webpage to the user, which is an extension to the relevance formula that was proposed earlier. The current work aims to create user profiles automatically and implicitly depending on the various web pages a user browses over a period of time and the user’s interaction with them. This automatically generated user profile assigns weights to web pages proportional to the user interactions on the webpage and thus indicates relevancy of web pages to the user based on these weights. © Springer International Publishing Switzerland 2016.