System and method for concepts caching using a deep-content-classification (dcc) system

Abstract

A method and system for caching concept structures in a cache memory of a user computing device are provided. The method includes collecting at least one environmental variable related to a user of the computing device; retrieving at least one concept structure matching the at least one received environmental variable, each concept structure comprises a concept metadata associated with a plurality of multimedia content elements; and caching the at least one matching concept structure in the cache memory.

We claim: 
     
         1 . A method for reducing bandwidth consumption by a user computing device, comprising:
 collecting at least one environmental variable related to current environment of the user computing device;   attempting to match the at least one environmental variable to at least one concept structure from a plurality of concept structures stored on another computing device, wherein each concept structure in the plurality of concept structures comprises a concept metadata and an associated reduced cluster of signatures derived from a plurality of multimedia content elements;   upon determining that the attempting to match was successful, retrieving the at least one concept structure;   caching the at least one concept structure in the cache memory receiving an input multimedia content element in a query;   using the at least one concept structure in the cache memory to perform local analysis of the input multimedia content element; and   providing a response to the query based on the local analysis.   
     
     
         2 . The method of  claim 1 , wherein the environmental variable includes at least one of: a time of day, location information, motion information, and weather information. 
     
     
         3 . The method of  claim 2 , wherein the at least one environmental variable is collected by at least one sensor of the computing device. 
     
     
         4 . The method of  claim 3 , wherein the at least one sensor is any one of: a camera, a web camera, a microphone, a Global Positioning System, an image analyzer, and a speech recognizer. 
     
     
         5 . The method of  claim 1 , wherein the retrieving comprises fetching the at least one concept structure from at least one of: a deep-content-classification system or a data warehouse. 
     
     
         6 . The method of  claim 1 , wherein the attempting to match comprises:
 deriving an environmental signature from the environmental variable;   comparing the environmental signature to the associated reduced cluster of signatures; and   determining that the at least one concept structure matches the at least one environmental variable according to an overlap of more than a predefined threshold between the environmental signature and at least one signature in the associated reduced cluster of signatures.   
     
     
         7 . The method of  claim 1 , wherein the at least one concept structure is cached on the computing device for a predefined period of time set by an aging timer. 
     
     
         8 . The method of  claim 1 , wherein the local analysis comprises: determining input metadata associated with the input multimedia content element; and matching the at least one multimedia content element to the at least one concept structure cached in the cache memory. 
     
     
         9 . The method of  claim 8 , wherein the determining input metadata comprises at least one of: identifying metadata associated with the input multimedia content element; or generating metadata based on environmental variables collected upon the receiving of the input multimedia content element. 
     
     
         10 . The method of  claim 1 , wherein the concept structure is determined to match the input multimedia content element when the input metadata overlaps the concept metadata in the concept structure. 
     
     
         11 . The method of  claim 1 , wherein the input multimedia content element is at least one of: an image, graphics, a video stream, a video clip, an audio stream, an audio clip, a video frame, a photograph, images of signals, medical signals, geophysical signals, subsonic signals, supersonic signals, electromagnetic signals, and infrared signals. 
     
     
         12 . The method of  claim 1 , wherein the method is performed locally by the computing device. 
     
     
         13 . The method according to  claim 1  comprising applying a periodic cluster reduction on each signature reduced cluster. 
     
     
         14 . The method according to  claim 1  comprising generating each one of the signatures by statistically independent computational cores. 
     
     
         15 . The method according to  claim 1  wherein the environmental variable includes weather information. 
     
     
         16 . The method according to  claim 1  wherein the environmental variable includes motion information. 
     
     
         17 . The method according to  claim 1  comprising generating the plurality of concept structures in an automatic and unsupervised manner. 
     
     
         18 . The method according to  claim 1  wherein each cluster concept structure in the plurality of concept structures does not include any of the plurality of multimedia content elements. 
     
     
         19 . A non-transitory computer readable medium having stored thereon instructions for causing one or more processing units to execute the following method steps:
 collecting at least one environmental variable related to a current environment of a computing device;   attempting to match the at least one environmental variable to at least one concept structure from a plurality of concept structures, wherein each concept structure in the plurality of concept structures comprises a concept metadata and an associated reduced cluster of signatures derived from a plurality of multimedia content elements;   upon determining that the attempting to match was successful, retrieving the at least one concept structure;   caching the at least one concept structure in the cache memory;   receiving an input multimedia content element in a query;   using the at least one concept structure in the cache memory to perform local analysis of the input multimedia content element; and   providing a response to the query based on the local analysis.   
     
     
         20 . A system for caching concept structures in a cache memory of a user computing device, comprising:
 a processing unit; and   a memory connected to the processing unit, wherein the memory contains instructions that when executed by the processing unit, configure the system to:   receive at least one environmental variable related to a current environment of the user computing device;   attempt to match the at least one received environmental variable to at least one concept structure from a plurality of concept structures, wherein each concept structure in the plurality of concept structures comprises a concept metadata and an associated reduced cluster of signatures derived from a plurality of multimedia content elements;   determine that the attempt to match was a success match;   for a successful match, retrieve the at least one concept structure;   cache the at least one concept structure in the cache memory;   receive an input multimedia content element in a query;   use the at least one concept structure in the cache memory to perform local analysis of the input multimedia content element; and   provide a response to the query based on the local analysis.