Abstract: The present invention relates to a television system (50) and a method for automatically suggesting suitable programs to a viewer from a large number of available programs. The system (50) includes a DTV-agent system (21). Title information and characteristics of programs are made available as EPG (Electronic Program Guide) data, which including at least one Electronic Program Guide Database (22). A learning module (39) records characteristics associated with each program viewed by the user, and forms sets of these characteristics. The frequency of occurrence of each set is also determined. A recommendation module (40) uses a number of tasks to compile a list of viewer recommendations (67). Various tasks are defined, with each task defining a unique combination of a manner of ordering the viewer profile (500), and particular Relevance Filters for filtering the ordered viewer profile.

Abstract: A system (50) for earning functional usage patterns of a user of the system (50) and for planning a best sequences of events suitable for a particular user in a particular context is disclosed. Information relating to user decisions, such as the context during which the decision was made and the actual user decision, is gathered by the DTV-agent (36) and delivered to the active avatar agent (37). The learning module (39) operates to identify all generalization patterns from a number of instances. Method (500) then determines which particular decision is most appropriate by comparing the generalization patterns with the current context for which a decision must be made. Clusters of such behavior patterns are formed with each cluster having the same number of matched attributes. A decision selecting process then selects a decision which is most appropriate to the current context from the behavior patterns.

Abstract: A system (50) for earning functional usage patterns of a user of the system (50) and for planning a best sequences of events suitable for a particular user in a particular context is disclosed. Information relating to user decisions, such as the context during which the decision was made and the actual user decision, is gathered by the DTV-agent (36) and delivered to the active avatar agent (37). The learning module (39) operates to identify all generalisation patterns from a number of instances. Method (500) then determines which particular decision is most appropriate by comparing the generalisation patterns with the current context for which a decision must be made. Clusters of such behaviour patterns are formed with each cluster having the same number of matched attributes. A decision selecting process then selects a decision which is most appropriate to the current context from the behaviour patterns.

Abstract: A method (1800) for controlling a system comprising a plurality of appliances, each said appliance having a corresponding appliance agent, and a user agent, associated with a user is disclosed. The method detecting (1802) an interaction of the user with a corresponding first appliance, communicating (1804) this to the user agent to thereby update a user profile, sending (1808) a first action message dependent upon the user profile and upon said one of the interaction to the first appliance agent to direct the corresponding first appliance to perform a first action, wherein control over said first action thereby passes from the user agent to the first appliance agent.