Subscriber characterization system with filters

Abstract

A subscriber characterization system with filters in which the subscriber's selections are monitored, including monitoring of the time duration programming is watched, the volume at which the programming is listened to, and any available information regarding the type of programming, including category and sub-category of the programming. The raw subscriber selection data is then processed to eliminate data associated with irrelevant activities such as channel surfing, channel jumping, or extended periods of inactivity. The actual subscriber selection data is used to form program characteristics vectors. The programming characteristics vectors can be used in combination with the actual subscriber selection data to form a subscriber profile. Heuristic rules indicating the relationships between programming choices and demographics can be applied to generate additional probabilistic subscriber profiles regarding demographics and programming and product interest.

What is claimed is: 
     
         1 . A computer-implemented method of inferring at least one characteristic of a viewer of multimedia programming, the method comprising:
 obtaining viewer selection data having source material, times, and Electronic Program Guide data of the viewer;   processing the viewer selection data to obtain filtered viewer selection data, wherein processing includes applying at least one filter to the viewer selection data to exclude viewer selection data that was associated with a time occurring within one or more particular periods of time;   defining one or more heuristic rules that relate at least one aspect of the filtered viewer interaction data to at least one characteristic of the viewer; and   applying one or more of the heuristically developed rules to at least a subset of the filtered viewer selection data to infer at least one characteristic of the viewer from the filtered viewer selection data.   
     
     
         2 . The method of  claim 1 , wherein one of the at least one inferred characteristics is an inferred age. 
     
     
         3 . The method of  claim 2 , wherein the inferred age is an age range. 
     
     
         4 . The method of  claim 1 , wherein the heuristically developed rules are probabilistic in nature. 
     
     
         5 . The method of  claim 1 , wherein each of the at least one inferred characteristics is expressed as a probability assigned by the rules based on the filtered viewer interaction data. 
     
     
         6 . The method of  claim 1 , wherein the viewer selection data includes data describing at least some subset of channel changes, volume changes, record commands, and time of viewer interaction. 
     
     
         7 . The method of  claim 1 , wherein the processing includes evaluating channel change commands and associated viewing times to determine the filtered viewer interaction data. 
     
     
         8 . The method of  claim 1 , wherein the one are more heuristic rules include an inferential link between the filtered viewer interaction data and at least one of a plurality of potential viewer characteristics. 
     
     
         9 . The method of  claim 2 , wherein the application of the heuristically developed rules provides a predictive value of the inferred age. 
     
     
         10 . The method of  claim 1 , wherein Electronic Programming Guide includes programming category, program description, rating, actors, and program duration. 
     
     
         11 . The method of  claim 1 , wherein source material includes analog video, (MPEG) digital video source material, Hypertext Markup Language (HTML), multimedia source material, source related textual information having descriptive fields which are related to the source and text which is part of the source material. 
     
     
         12 . A computer-implemented method of inferring at least one characteristic of a user of a multimedia device, the method comprising:
 monitoring user interactions of the multimedia device, to obtain user selection data that includes source material, times, and Electronic Program Guide data of the user;   processing the user selection data to obtain filtered user selection data, wherein processing includes applying at least one filter to the user selection data to exclude user selection data that was associated with a time occurring within one or more particular periods of time;   defining one or more heuristic rules that relate at least one aspect of the filtered user interaction data to at least one characteristic of the user; and   applying one or more of the heuristically developed rules to at least a subset of the filtered user selection data to infer at least one characteristic of the user from the filtered user selection data.   
     
     
         13 . The method of  claim 12 , wherein one of the at least one inferred characteristics is an inferred age. 
     
     
         14 . The method of  claim 13 , wherein the inferred age is an age range. 
     
     
         15 . The method of  claim 12 , wherein the heuristically developed rules are probabilistic in nature. 
     
     
         16 . The method of  claim 12 , wherein each of the at least one inferred characteristics is expressed as a probability assigned by the rules based on the filtered user interaction data. 
     
     
         17 . The method of  claim 12 , wherein the user selection data includes data describing at least some subset of channel changes, volume changes, record commands, and time of user interaction. 
     
     
         18 . The method of  claim 12 , wherein the processing includes evaluating channel change commands and associated viewing times to determine the filtered user interaction data. 
     
     
         19 . The method of  claim 13 , wherein the application of the heuristically developed rules provides a predictive value of the inferred age. 
     
     
         20 . A viewer profiling system in a network environment for determining at least one characteristic of a viewer of multimedia, the system comprising:
 a viewer selection module configured to monitor interaction of a viewer of multimedia and to process the viewer selection data by applying at least one filter to the viewer selection data to exclude viewer selection data that was associated with a time occurring within one or more particular periods of time;   a storage module for storing one or more heuristic rules, wherein the one or more heuristic rules relate at least one aspect of the filtered viewer interaction data to at least one characteristic of the viewer;   a rule application module configured to retrieve one or more heuristic rules applicable to the filtered viewer selection data, to apply the one or more retrieved heuristic rules to at least a subset of the filtered viewer selection data and to infer at least one characteristic of the viewer from the filtered viewer selection data.