Method and system for automatically determining demographics of media assets for targeting advertisements

1. A computer-implemented method of matching an advertisement for a service to a television media asset comprising:

extracting, by a processor, at least one service keyword for a service from a web page associated with the service;

generating, by the processor, a service profile for the service comprising the at least one extracted service keyword for the service;

generating, by the processor, a plurality of media asset profiles corresponding to a plurality of television media assets, each media asset profile including at least one media asset keyword for a corresponding television media asset;

calculating, for each television media asset, a match score based on the at least one media asset keyword and the at least one extracted service keyword;

calculating, for the service profile, a plurality of fitness scores based at least in part on the match score for each television media asset; and

determining a particular television media asset for placing an advertisement for the service based upon the plurality of fitness scores.

2. The method of claim 1 , further comprising one of the following:

showing the fitness scores as part of a media asset schedule; showing the fitness scores and upper and lower confidence bounds of the fitness scores as part of a media asset schedule; showing the fitness scores superimposed on the media asset schedule; and showing forecast fitness scores and actual fitness scores on the media fitness schedule.

3. The method of claim 1 , wherein the service profile further comprises a service demographic profile for the service based on customers of the service, wherein each media asset profile further comprises a media demographic profile for a television media asset based upon viewer demographic information, and wherein the fitness scores are calculated based upon the match scores and upon demographic similarity measures between the service demographic profile for the service and the media demographic profiles for the plurality of television media assets.

4. The method of claim 3 , wherein the service demographic profile for the service and the media demographic profiles for the plurality of television media assets are weighted differentially.

5. The method of claim 3 , further comprising transforming the media demographic profiles of the plurality of television media assets to be compatible with the service demographic profile, or transforming the service demographic profile of the service to be compatible with the media demographic profiles of the plurality of television media assets, wherein information for determining the plurality of media demographic profiles and the service demographic profile come from different sources.

6. The method of claim 1 , wherein the fitness scores are further based on at least one of historical media performances between the plurality of television media assets and the service or historical media costs for the plurality of television media assets.

7. The method of claim 1 , further comprising:

automatically extracting the one or more media asset keywords for the plurality of television media assets, wherein the media asset keywords for a television media asset are extracted from at least one of closed captioning text associated with the television media asset, a program schedule description associated with the television media asset, or speech-to-text conversion for program content of the television media asset.

8. The method of claim 3 , wherein generating the service profile for the service comprises:

automatically selecting previous customers of the service using a plurality of customer records from previous service orders for the service;

concatenating a plurality of enriched customer records for the previous customers based on adding customer demographic information to the plurality of customer records, wherein the customer demographic information comprises a plurality of variables; and

creating, by the processing device, the service demographic profile for the service from the plurality of enriched customer records, the service demographic profile comprising automatically scored target demographics for the service based at least in part on a combination of the customer demographic information from the plurality of enriched customer records.

9. The method of claim 3 , further comprising providing most significant matching or most significant mis-matching demographics between the service demographic profile and the media demographic profiles.

10. The method of claim 3 , further comprising:

assigning a unique identification code to each of a plurality of media asset fitness models;

testing media asset fitness models of the plurality of media asset fitness models against other media asset fitness models; and

tracking results of the testing using the unique identification codes.

11. The method of claim 1 , wherein the fitness scores are further based upon a demographic similarity determined by calculating a function based on a first demographic of customers of the service and a second demographic of linked television customers, wherein one or more linking keys permit television viewing activities to be tracked.

12. The method of claim 11 , wherein the linked television customers bought the service by one of the following methods: calling a phone number that was provided by a television advertisement; accessing a uniform resource locator (URL) that was provided by a television advertisement; or redeeming a special offer that was provided by a television advertisement.

13. The method of claim 11 , wherein demographic information about the linked television customers is automatically appended with additional information for use in determining similarity.

14. The method of claim 1 , wherein the fitness scores are further based upon a demographic similarity determined by calculating a function based on a first demographic of customers of the service and a second demographic of a television viewing panel, wherein the television viewing panel permits television viewing activities to be tracked.

15. The method of claim 1 , wherein the plurality of television media assets are selected from the following: a television station, a television media asset in a particular geographic location, a television media asset program, a television media asset on a particular day, a television media asset running at a particular hour, or a television media asset running at a particular minute.

16. The method of claim 1 , further comprising estimating a value of a television media asset as an advertisement vehicle for bidding on an auction for advertising with the television media asset.

17. A computer-implemented method of matching an advertisement for a service to a television media asset comprising:

extracting, by a processor, one or more media asset keywords for a plurality of television media assets;

generating a plurality of media asset profiles for the plurality of television media assets, each media asset profile for a television media asset comprising the one or more media asset keywords for the television media asset;

generating, by the processor, a service profile for a service, the service profile comprising one or more service keywords for the service;

calculating, for each television media asset, a match score between the one or more media asset keywords for the television media asset and the one or more service keywords for the service;

calculating fitness scores between the service profile for the service and the plurality of media asset profiles for the plurality of television media assets based at least in part on the match scores; and

selecting a particular television media asset for placing the advertisement for the service based upon the fitness scores.

18. The method of claim 17 , wherein the one or more television media assets comprise one of the following: short-form airings, long-form airings, a television station, a television program, a television program category, a day, an hour, a minute, a date time, program keywords, airing in a geographical location or market, or a television station and a day and hour combination.

19. The method of claim 17 , further comprising one of the following: showing the fitness scores as part of a media asset schedule; showing the fitness scores and upper and lower confidence bounds of the fitness scores as part of a media asset schedule; showing the fitness scores superimposed on the media asset schedule; and showing the forecast fitness scores and actual fitness scores on the media fitness schedule.

20. A non-transitory computer-readable storage medium for matching an advertisement for a service to a television media asset, the non-transitory computer-readable medium having instructions that, when executed by a processor, cause the processor to perform the operations comprising:

extracting at least one service keyword for a service from a web page associated with the service;

generating a service profile for the service comprising the at least one extracted service keyword for the service;

generating a plurality of media asset profiles corresponding to a plurality of television media assets, each media asset profile including at least one media asset keyword for a corresponding television media asset;

calculating, for each television media asset, a match score based on the at least one media asset keyword and the at least one extracted service keyword;

calculating, for the service profile, a plurality of fitness scores based at least in part on the match score for each television media asset; and

determining a particular television media asset for placing an advertisement for the service based upon the plurality of fitness scores.