System for recommending favorite channel/program based on TV watching pattern and method thereof
AssigneeElectronics and Telecommunications Research Institute
InventorJungwoo Joh
Life of this patent
Estimating the term…
Abstract
Disclosed are a system for recommending a favorite channel/program based on a television (TV) watching pattern and a method thereof. The present invention provides a system and a method for recommending a favorite channel/program by learning a television (TV) watching pattern using a Bayesian network, by extracting recommended candidate favorite channel/programs by applying an entropy theory to the learned Bayesian network, and by applying a fuzzy logic operation to the extracted candidate favorite channel/programs.
Claims14
A system for recommending a favorite channel/program, the system comprising:
- a processor;
- and
- a non-transitory computer-readable medium having program instructions embodied therewith, the program instructions being executable by the processor to function as
- a Bayesian network learning unit configured to learn a Bayesian network based on received watching pattern information of a user and to thereby calculate a first probability value with respect to all of the channels or programs based on the learning result;
- a candidate favorite channel/program extracting unit configured to calculate an entropy of the calculated first probability value, and to extract, when the entropy is less than or equal to a predetermined threshold, any of the channels or programs that corresponds to the calculated first probability value as a candidate favorite channel or program;
- a fuzzy logic operation unit configured to perform a fuzzy logic inference operation with respect to the extracted candidate favorite channel or program, and to calculate a second probability value with respect to each candidate favorite channel or program based on a result of the fuzzy logic inference operation;
- and
- a favorite channel/program recommending unit configured to determine a favorite channel or program according to the watching pattern information of the user based on the calculated first probability value and second probability value.
A computer-executable method of recommending a favorite channel/program, the method comprising:
- learning, by a processor, a Bayesian network based on received watching pattern information of a user to thereby calculate a first probability value with respect to all of the channels or programs based on the learning result;
- calculating, by the processor, an entropy of the calculated first probability value, and extracting, when the entropy is less than or equal to a predetermined threshold, any of the channels or programs that corresponds to the calculated first probability value as a candidate favorite channel or program;
- performing, by the processor, a fuzzy logic inference operation with respect to the extracted candidate favorite channel or program, and calculating a second probability value with respect to each candidate favorite channel or program based on a result of the fuzzy logic inference operation;
- and
- determining, by the processor, a favorite channel or program according to the watching pattern information of the user based on the calculated first probability value and second probability value.
Dependencies are inferred from the claim text.