Abstract
We present a method to extract implicit data of registered users of a web site with the help of mouse tracking. This allows us to generate more accurate interest profiles of visitors of a web site and to obtain a solid basis for the calculation of user interests or trend detection in the Web. On the one hand, web site owners have the opportunity to adjust their sites to the interests of their users. On the other hand, they can detect new trends and expand these topics on their web pages accordingly.