System and method for generating metadata for programming events

A system for generating metadata describing a video program, the system comprising:

• a programmable device including a computer readable medium storing programming code to control the device to perform processing comprising:
• obtaining, by a programmable device, production data corresponding to the video program from a production system used in the production of the video program;
• assigning, by the programmable device, respective numerical goodness of fit scores to respective predefined categories based on analysis of the production data to describe the subject matter of the video program, wherein the numerical goodness of fit score assigned to a category represents a degree to which the category is descriptive of the subject matter of the video program;
• assigning, by the programmable device, keywords to the video program based on analysis of the production data;
• storing, by the programmable device, numerical goodness of fit scores and keywords for the video program in association with time data and descriptive data for the video program as metadata describing the video program;
• and
• wherein said predefined categories are subject matter categories arranged in a hierarchy comprising at least a set of top-level categories, respective sets of first level sub-categories each corresponding and encompassed by a top level category, and respective sets of second level sub-categories each corresponding to and encompassed by a first level sub-category.

A method in a programmable device for generating metadata describing a programming event, the method comprising:

• obtaining production data corresponding to the programming event from a production system used in the production of the programming event, the production data including descriptive information for the programming event;
• determining candidate keywords from the production data;
• providing the candidate keywords as respective inputs to a classification tool and generating for each of said candidate keywords a set of numerical goodness of fit scores each corresponding to a predefined subject matter category, wherein the numerical goodness of fit score corresponding to a category represents a degree to which the category is descriptive of the candidate keyword;
• selecting keywords to represent the programming event from among said candidate keywords based on the set of numerical goodness of fit scores for each of said candidate keywords;
• and
• storing said selected keywords in a computer readable medium as a component of said metadata describing the programming event;
• and
• wherein said predefined categories are subject matter categories arranged in a hierarchy comprising at least a set of top level categories, respective sets of first level-categories each corresponding and encompassed by a top level category, and respective sets of second level sub-categories each corresponding to and encompassed by a first level sub-category.

A programmable device for generating metadata for transmission to a programming event receiver, the metadata describing a programming event, the device comprising a computer readable medium storing programming code for controlling the device to perform processing comprising:

• obtaining production data corresponding to the programming event from a production system used in the production of the programming event, the production data including descriptive information for the programming event;
• determining candidate keywords from the production data;
• providing the candidate keywords as respective inputs to a classification tool and generating for each of said candidate keywords a set of numerical goodness of fit scores each corresponding to a predefined subject matter category, wherein the numerical goodness of fit score corresponding to a category represents a degree to which the category is descriptive of the candidate keyword;
• selecting keywords to represent the programming event from among said candidate keywords based on the set of numerical goodness of fit scores for each of said candidate keywords;
• and
• storing said selected keywords in a computer readable medium as a component of said metadata describing the programming event;
• and
• wherein said predefined categories are subject matter categories arranged in a hierarchy comprising at least a set of top-level categories, respective sets of first level sub-categories each corresponding to and encompassed by a top level category, and respective sets of second level sub-categories each corresponding to and encompassed by a first level sub-category.

A method for generating metadata describing the subject matter of individual segments of a video program, the method comprising:

• obtaining, by a programmable device, production data corresponding to the video program from a production system used in the production of the video program;
• processing, by the programmable device, the production data to determine individual segments of the video program prior to broadcast of the video program;
• for each segment of the video program, the programmable device:
• assigning respective numerical goodness of fit scores to respective predefined categories based on analysis of the production data to describe the subject matter of the segment of the video program, wherein the numerical goodness of fit score assigned to a category represents a degree to which the category is descriptive of the subject matter of the segment of the video program;
• assigning keywords to the segment of the video program based on analysis of the production data;
• and
• storing numerical goodness of fit scores and keywords for the segment of the video program in a computer readable medium in association with time data and descriptive data for the segment of the video program as metadata describing the segment of the video program;
• and
• wherein said predefined categories are subject matter categories arranged in a hierarchy comprising at least a set of top-level categories, respective sets of first level sub-categories each corresponding to and encompassed by a top level category, and respective sets of second level sub-categories each corresponding to and encompassed by a first level sub-category.

A system for generating metadata describing the subject matter of individual segments of a video program, the system comprising:

• a programmable device including a computer readable medium storing programming code to control the programmable device to perform processing comprising:
• obtaining, by a programmable device, production data corresponding to the video program from a production system used in the production of the video program;
• processing, by the programmable device, the production data to determine individual segments of the video program prior to broadcast of the video program;
• for each segment of the video program, the programmable device:
• assigning respective numerical goodness of fit scores to respective predefined categories based on analysis of the production data to describe the subject matter of the segment of the video program, wherein the numerical goodness of fit score assigned to a category represents a degree to which the category is descriptive of the subject matter of the segment of the video program;
• assigning keywords to the segment of the video program based on analysis of the production data;
• and
• storing numerical goodness of fit scores and keywords for the segment of the video program in a computer readable medium in association with time data and descriptive data for the segment of the video program as metadata describing the segment of the video program;
• and
• wherein said predefined categories are subject matter categories arranged in a hierarchy comprising at least a set of top-level categories, and respective sets of first level sub-categories each corresponding to and encompassed by a top level category, and respective sets of second level sub-categories each corresponding to and encompassed by a first level sub-category.