Advertisement monitoring system
View Patent ↗An advertising monitoring system is presented in which subscriber selections including channel changes are monitored, and in which information regarding an advertisement is extracted from text related to the advertisement. The text related to the advertisement is in the form of closed caption text, data transmitted with the advertisement, or other associated text. A record of the effectiveness of the advertisement is created in which measurements of the percentage of the advertisement which was viewed are stored. Such records allow a manufacturer or advertiser to determine if their advertisement is being watched by subscribers. The system can be realized in a client-sever mode in which subscriber selection requests are transmitted to a server for fulfillment, in which case the advertisement monitoring takes place at the server side.
1. A computer-implemented method of inferring the gender of a consumer, the method comprising:
(a) collecting closed captioning data associated with a television program being presented to a consumer;
(b) retrieving at least one keyword and a probabilistic measure of gender associated with the at least one keyword;
(c) evaluating the closed captioning data for the at least one keyword; and
(d) inferring the gender of the consumer based on the probabilistic measure of gender associated with the keywords evaluated in step (c).
2. The method of claim 1 , wherein the at least one keyword is obtained through the use of one or more pre-defined heuristic rules, wherein the pre-defined heuristic rules relate gender to keywords contained in television programs.
3. The method of claim 1 , further comprising:
(f) aggregating the closed captioning data with previously stored closed captioning data;
(g) inferring at least one consumer demographic characteristic based on the aggregated closed captioning data; and
(h) displaying advertisements to the consumer based on the at least one demographic characteristic of step (g).
4. The method of claim 1 , wherein the evaluating of step (c) is accomplished by comparing the at least one keyword to words in the closed captioning data.
5. The method of claim 1 , wherein the evaluating of step (c) includes comparing synonyms of the at least one keyword to words in the closed captioning data.
6. The method of claim 1 , further comprising:
selecting an advertisement for presentation to the consumer based at least in part on the inferred gender.
7. A computer-implemented method of inferring the gender of a subscriber based on the subscriber's interaction with targeted programming, the method comprising:
(a) monitoring the subscriber's viewing habits in a television system;
(b) retrieving information associated with programs viewed by the subscriber, the information including a program description;
(c) applying a set of one or more pre-defined heuristic rules, wherein the set of pre-defined heuristic rules is selected based on the program descriptions associated with the subscriber's viewing habits;
(d) inferring at least one subscriber characteristic of the subscriber based on the application of the pre-defined heuristic rules;
(e) correlating the at least one subscriber characteristic with at least one characteristic of a gender; and
(f) associating the subscriber with the gender if there is a sufficient correlation between the at least one subscriber characteristic and the at least one characteristic of the gender.
8. The method of claim 7 , wherein the pre-defined heuristic rules assign probabilities of the subscriber having the at least one subscriber characteristic.
9. The method of claim 7 , wherein the at least one subscriber characteristic is based on subscriber interactions that are aggregated over multiple viewing sessions.
10. The method of claim 7 , wherein step (a) includes determining the extent to which the subscriber interacts with the programming.