Evaluation of web user interaction serves various purposes like analysis and evolution of UX, but it requires significant resources. To reduce the time and cost associated with such assessments, several automated solutions have been developed to analyze interactions by capturing logs. However, many of these approaches assume that all users interact similarly, disregarding individual characteristics such as mouse and keyboard movement speeds. Additionally, they often overlook the time required for analysis. We propose that identifying a user’s interaction profile can enhance the quality of automated log analysis. To achieve this, in this proposal interactive user profiles on the web will be studied and defined. Through experimentation, we will analyze real user behavior and its relationship with these profiles. Finally, the detection of user profiles and interactive characteristics will be automated, considering key variables such as detection speed and prediction accuracy.