No matter which construction method is chosen, the profile must be kept current to reflect the user’s preferences accurately; this has proven to be a very challenging task.
Unlike the majority of known personalized search systems, TaskSieve aims to support the task-based exploratory search process. In place of a traditional model of user interests, TaskSieve applies a more focused task model, which attempts to accumulate information about the task explored by the user.
TaskSieve is unique because of its several innovative features. It aims to integrate users short-term interests (as queries) with their relative long-term task characteristics and preference (as the task model) to cope with the multiple iterations of the exploration of search space. The system also subjects the integration under the users control through a set of predefined combination modes, so that the system and the process are more flexible and transparent. As the second innovative feature, TaskSieve returns documents surrogates that are task-infused by the generation of their content and by the highlighting of terms within them. This gives the users more direct clues about the potential relevance of the documents to not only their queries, but also the task model. Finally, TaskSieve also provides on-screen visualization of the task model as the third innovation feature so that such information is always available to the users during all their searches.
In user customization, a recommendation system provides an interface that allows users to construct a representation of their own interests.
Content-based recommendation systems recommend an item to a user based upon a description of the item and a profile of the user’s interests. While a user profile may be entered by the user, it is commonly learned from feedback the user provides on items. A variety of learning algorithms have been adapted to learning user profiles, and the choice of learning algorithm depends upon the representation of content.
No comments:
Post a Comment