Distributed and tiered architecture for content search and content monitoring
View Patent ↗An efficient large scale search system for video and multi-media content using a distributed database and search, and tiered search servers is described. Selected content is stored at the distributed local database and tier1 search server(s). Content matching frequent queries, and frequent unidentified queries are cached at various levels in the search system. Content is classified using feature descriptors and geographical aspects, at feature level and in time segments. Queries not identified at clients and tier1 search server(s) are queried against tier2 or lower search server(s). Search servers use classification and geographical partitioning to reduce search cost. Methods for content tracking and local content searching are executed on clients. The client performs local search, monitoring and/or tracking of the query content with the reference content and local search with a database of reference fingerprints. This shifts the content search workload from central servers to the distributed monitoring clients.
1. A method for content monitoring, the method comprising:
organizing a reference database into a first tier database having search caches of popular and real time live content and long time storage of the popular and real time content and a second tier database having content classified in separate databases;
generating fingerprints in a user device for user watched content that includes popular and real time live content;
searching for the user watched content in a reference database;
generating a menu of option based on a reference database content that matched with the user watched content;
searching in the search caches for the user watched content and returning to the user device the menu of options if a match is found in the search caches;
searching in the first tier long time storage if the user watched content is not found in the search caches and returning to the user device the menu of options if a match is found in the long time storage;
searching in the second tier database in a classified separate database according to a class of the user watched content if the user watched content is not found in the long time storage and returning to the user device the menu of options if a match is found in the second tier database; and
displaying the menu of options on the user device to solicit user selection of one displayed option.
2. The method of claim 1 , wherein the search caches comprise:
a found query cache storing content that matches frequent queries; and
an unidentified query cache storing queries that are not identified in the reference database.
3. The method of claim 1 , wherein the user watched content is not found in the reference database and the user watched content is stored in the reference database according to a class of the user watched content.
4. The method of claim 1 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
downloading information concerning the user watched content from a url specified with the action parameter.
5. The method of claim 1 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
controlling the user device based on the action parameter.
6. The method of claim 1 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
skipping to a particular time location in the user watched content according to a time parameter and the action parameter.
7. A method of performing search, the method comprising:
storing selected subsets of reference databases containing classified reference contents in a local cached database on a user device;
classifying content of a query generated on the user device to generate a classified query;
counting signatures of the classified query that are separated from other signatures in the classified query by a minimum distance to generate a query unique signature count;
performing a search with the classified query upon determining the query unique signature count exceeded a predetermined threshold;
comparing the classified query with the classified reference contents stored in the local cached database on the user device to perform the search with the classified query;
generating a menu of option based on a reference database content that matched from the search with the classified query; and
displaying the menu of options on the user device to solicit user selection of one displayed option.
8. The method of claim 7 further comprising:
determining to search a classified search database in a centralized server in response to determining the classified query did not find a match in the local cached database.
9. The method of claim 8 , wherein the classified search database is located on a search tier of a centralized server having a plurality of search tiers.
10. A method for content monitoring, the method comprising:
organizing a reference database into a first tier database having search caches of popular and real time live content;
generating fingerprints in a user device for user watched content that includes popular and real time live content;
searching in the search caches for the fingerprints for the user watched content and returning to the user device a menu of options if a match is found in the search cache; and
displaying the menu of options on the user device to solicit user selection of one displayed option.
11. The method of claim 10 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
downloading information concerning the user watched content from a url specified with the action parameter.
12. The method of claim 10 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
controlling the user device based on the action parameter.
13. The method of claim 10 further comprising:
responding to a user selection of the one option according to an action parameter stored with the selected menu option in the user device;
skipping to a particular time location in the user watched content according to a time parameter and the action parameter.
14. A method of performing search, the method comprising:
storing selected subsets of reference databases containing classified reference contents in a local cached database on a user device;
classifying content of a query generated on the user device to generate a classified query;
determining the information content of the classified query by using an entropy based information measurement on the classified query content;
performing a search with the classified query upon determining the entropy based information measured exceeds a predetermined threshold;
comparing the classified query with the classified reference contents stored in the local cached database on the user device to perform the search with the classified query;
generating a menu of option based on a reference database content that matched from the search with the classified query; and
displaying the menu of options on the user device to solicit user selection of one displayed option.
15. The method of claim 14 further comprising:
determining to search a classified search database in a centralized server in response to determining the classified query did not find a match in the local cached database.
16. The method of claim 15 , wherein the classified search database is located on a search tier of a centralized server having a plurality of search tiers.