System and method for selecting a video for insertion into an online web page
A system for selecting a video to be inserted into an article displayed at a web page includes a blackboard. The blackboard stores extracted values for operation thereon by one or more workers. An article extractor creates an article feature as a function of information extracted from the article and inputs the article feature to the blackboard. A threshold extractor determines a threshold factor as a function of the relevance of the video to the article, and inputting the threshold factor to the blackboard. A more like this recommender receives the article feature and outputs to the blackboard, as a function of the article feature, a more like this recommendation as a list of videos to be placed in the article. A more like this decider determines, as a function of the threshold factor and the more like this recommendation, which video to select as a more like this decision.
1. A system for selecting a video to be inserted into an article displayed at a web page comprising:
a processor;
a blackboard for storing extracted values for operation thereon by the processor, the processor including one or more workers; the workers including:
an article extractor, extracting information from the article, communicating with the blackboard, and creating an article feature as a function of information extracted from the article and inputting the article feature to the blackboard;
a threshold extractor, communicating with the blackboard, determining a threshold factor from a dynamo table, the threshold factor indicating a predetermined threshold of relevance of the video to the article, and inputting the threshold factor to the blackboard;
a more like this recommender, communicating with the blackboard, receiving the article feature from the blackboard and generating and outputting to the blackboard, as a function of the article feature, a more like this recommendation as a list of videos to be placed in the article and inputting the more like this recommendation to the blackboard; and
a more like this decider, communicating with the blackboard, and determining, as a function of the threshold factor and the more like this recommendation, which video from the list of videos to select to be inserted into the article.
2. The system for selecting a video of claim 1 , wherein the article feature includes at least one of a title of an article and a text of an article.
3. The system for selecting a video of claim 2 , wherein the article extractor communicates with a page service to obtain the title and the text of the web page at which the video is being placed.
4. The system for selecting a video of claim 1 , further comprising an endpoint for initiating a request to place a video by inputting an initial request to the blackboard, the article extractor creating an article feature in response to the initial request.
5. The system for selecting a video of claim 4 , wherein the initial request includes a URL identifier, and at least one of user information, a user ID, the title of the article located at the URL, and an identifier for tracking interactions by a particular user and the webpage associated with the URL.
6. The system for selecting a video of claim 4 , wherein the threshold extractor determines a threshold factor in response to the initial request.
7. The system for selecting a video of claim 1 , wherein the more like this recommender determines a more like this recommendation by comparing a frequency of uncommon words appearing in the article and a frequency of the uncommon words appearing in a description of each video of the list of videos.
8. A computer program product for selecting a video to be inserted into an article displayed at a web page comprising:
a non transitory computer readable storage medium and program instructions stored on the computer readable storage medium, the program instructions comprising instructions to:
store extracted values on a blackboard for operation thereon by one or more workers; extract an article feature from an article, wherein the article feature is input into the blackboard;
obtain a threshold factor from a dynamo table, wherein the threshold factor is input into the blackboard, the threshold factor indicating a predetermined threshold of relevance of the video to the article; utilize the article feature, to enable a more like this recommender to generate a list of videos to be placed in the article; and
utilize the threshold factor and the more like this recommender list, to enable a more like this decider to determine which video to select for insertion into the article.
9. The computer program product for selecting a video of claim 8 , wherein the article feature includes at least one of a title of an article and text of an article.
10. The computer program product for selecting a video of claim 9 , wherein extracting an article feature from an article requires communication with a page service to obtain the title and the text of the web page at which the video is being placed.
11. The computer program product for selecting a video of claim 8 , further comprising an endpoint for initiating an initial request to place a video on the webpage, to the blackboard, wherein the article feature is created in response to the initial request.
12. The computer program product for selecting a video of claim 11 , wherein the threshold factor is determined in response to the initial request.
13. The computer program product for selecting a video of claim 8 , wherein the more like this recommender determines a more like this recommendation list by comparing a frequency of uncommon words appearing in the article and a frequency of the uncommon words appearing in a description of each video of the list of videos.
14. A system for selecting a video to be inserted into an article displayed at a webpage comprising:
one or more computer processors;
one or more computer readable media; program instructions stored therein for execution by at least one or more computer processors, the program instructions comprising instructions to:
store extracted values on a blackboard for operation thereon by one or more workers; extract an article feature from an article, wherein the article feature is input into the blackboard;
determine a threshold factor from a dynamo table, wherein the threshold factor is input into the blackboard, the threshold factor indicating a predetermined threshold of relevance of the video to the article;
utilize the article feature, to enable a more like this recommender to generate a list of videos to be placed in the article; and
utilize the threshold factor and the more like this recommender list, to enable a more like this decider to determine which video to select for insertion into the article.
15. The system for selecting a video of claim 14 , wherein the article feature includes at least one title of an article and at least one text of an article.
16. The system for selecting a video of claim 15 , wherein extracting an article feature from an article requires communication with a page service to obtain the title and the text of the web page at which the video is being placed.
17. The system for selecting a video of claim 14 , further comprising an endpoint for initiating an initial request to place a video on the webpage, to the blackboard, wherein the initial request creates the article feature in response to the initial request.
18. The system for selecting a video of claim 17 , wherein the threshold factor is determined in response to the initial request.
19. The system for selecting a video of claim 14 , wherein the more like this recommender determines a more like this recommendation list by comparing a frequency of uncommon words appearing in the article and a frequency of the uncommon words appearing in a description of each video of the list of videos.
20. A method for selecting a video to be inserted into an article displayed at a web page comprising:
storing extracted values on a blackboard for operation thereon by one or more workers; extracting an article feature from an article, wherein the article feature is input into the blackboard;
determining a threshold factor from a dynamo table, wherein the threshold factor is input into the blackboard, the threshold factor indicating a predetermined threshold of relevance of the video to the article;
wherein, a more like this recommender, receives the article feature and generates a list of videos to be placed in the article; and
wherein a more like this decider receives the threshold factor and the more like this recommender list and determines which video to select for insertion into the article.
21. The method for selecting a video of claim 20 , wherein the article feature includes at least one title of an article and at least one text of an article.
22. The method for selecting a video of claim 21 , wherein extracting an article feature from an article requires communication with a page service to obtain the title and the text of the web page at which the video is being placed.
23. The method for selecting a video of claim 20 , further comprising an endpoint for initiating an initial request to place a video on the webpage, to the blackboard, wherein the article feature is created in response to the initial request.
24. The method for selecting a video of claim 23 , wherein the threshold factor is determined in response to the initial request.
25. The method for selecting a video of claim 20 , wherein the more like this recommender determines a more like this recommendation list by comparing a frequency of uncommon words appearing in the article and a frequency of the uncommon words appearing in a description of each video of the list of videos.