The volume of information users are exposed to on the web is overwhelming. To increase effectiveness of information delivery to users, providers employ personalization strategies. In a highly competitive environment, simplistic strategies do not suffice, and high-quality personalization is required. These can be based on users' decision making models. To build such models, we need to extract factors of direct influence on users' decision making. Personality factors are known to have this direct influence. They are stable over time and across situations, and they assist in predicting future behavior of individuals in a scientific way. In this paper, we introduce a novel methodology for extracting users' personality factors without holding any prior information on the users' behavior and, notably, without administering any psychological questionnaires. This allows us to build a designated model for each user or users' group, and in turn facilitates effective personalized information delivery. Copyright is held by the owner/author(s).