Behavioral Segmentation For Web Searching Users Using Self-Organizing Kohonen Maps

Knowing the behavior of Internet users is a main and a very important concern for Internet providers. Therefore, clustering Internet users according to their behavior would be very useful in order to satisfy users’ needs and to prepare appropriate offers. Web usage mining analyzes logs of web servers in order to discover such clusters and determining users’ behavior accordingly. Kohonen's Self Organizing Feature maps are combined with k-mean clustering to segment Internet users, especially searching users. This has been done through studying their usage behavior, the resulted segments are discussed and some suggestions are provided on the way search engines should deal with those users.