Database for contextual personalization

In recent years, research into user centric and personalized applications has focused on the utilization of contextual information about the situation in which the user is consuming the content item. However, there is no database suitable for the investigation of specific open issues of contextual information description and utilization available today. The reason for this are several known difficulties with user related contextual information acquisition. A description of a publicly available database for personalization and user adaptation including contextual information is given in the paper. The data was acquired from users watching movies and then providing contextual information about the event in addition to submitting ratings about the movie. Beside describing raw data, the paper outlines basic statistics and selected properties of potentially contextual variables. At the time of submission, the database included 12 contextual variables, more than 90 users, 950 items and 1600 ratings. Data was acquired using a dedicated web application which is still publicly available and the data acquisition is still in progress. Content items (movies) can be enhanced by content item metadata using publicly available databases.