Video-based Adaptive Recommendations

Abstract

Methods and systems for video-based adaptive recommendations deliver recommendations to users that are based upon inferences from user behaviors, and further, from interpretations of pictorial-based information associated with video content through the application of computer-implemented neural networks. The recommendations may refer to information that is interpreted from the pictorial-based information such as representations of physical objects. Natural language-based explanations for the delivered recommendations that explain the reasoning that is applied in making the recommendations may be provided to recommendation recipients.

What is claimed is: 
     
         1 . A method, comprising:
 accessing automatically a plurality of usage behaviors associated with one or more users of a computer-implemented system;   inferring automatically a preference of a user of the computer-implemented system from the plurality of usage behaviors;   accessing automatically video content, wherein the video content comprises pictorial-based information;   interpreting automatically the pictorial-based information by application of a computer-implemented neural network;   generating automatically a recommendation that is based upon the inferring of the preference and the interpreting of the pictorial information; and   delivering automatically the recommendation to the user.   
     
     
         2 . The method of  claim 1 , further comprising:
 inferring automatically the preference of the user of the computer-implemented system from the plurality of usage behaviors, wherein the inferring of the preference is based upon the application of a plurality of inference weightings that are determined in accordance with usage behavior priority rules that are applied to the plurality of usage behaviors.   
     
     
         3 . The method of  claim 1 , further comprising:
 generating automatically the recommendation, wherein the recommendation comprises information that is associated with the interpreting of the pictorial information.   
     
     
         4 . The method of  claim 3 , further comprising:
 generating automatically the recommendation, wherein the recommendation comprises the information that is associated with the interpreting of the pictorial information, wherein the information comprises a reference to a physical object.   
     
     
         5 . The method of  claim 4 , further comprising:
 generating automatically the recommendation, wherein the recommendation comprises the information that is associated with the interpreting of the pictorial information, wherein the information comprises the reference to the physical object, wherein the physical object comprises a person.   
     
     
         6 . The method of  claim 1 , further comprising:
 generating automatically the recommendation, wherein the recommendation is generated in response to an inferred focus of attention of the user with respect to the video content.   
     
     
         7 . The method of  claim 1 , further comprising:
 generating automatically a natural language-based explanation for the delivery of the recommendation, wherein the explanation comprises reasoning that is associated with the generating of the recommendation; and   delivering automatically the explanation to the user.   
     
     
         8 . A computer-implemented system comprising one or more processors configured to:
 access a plurality of usage behaviors associated with one or more users of the computer-implemented system;   infer a preference of a user of the computer-implemented system from the plurality of usage behaviors;   access video content, wherein the video content comprises pictorial-based information;   interpret the pictorial-based information by application of a computer-implemented neural network;   generate one or more recommendations that are based upon the inferring of the preference and the interpretation of the pictorial information; and   deliver at least one of the one or more recommendations to the user.   
     
     
         9 . The system of  claim 8  comprising the one or more processors, further configured to:
 infer the preference of the user of the computer-implemented system from the plurality of usage behaviors, wherein the inference of the preference is based upon the application of a plurality of inference weightings that are determined in accordance with usage behavior priority rules that are applied to the plurality of usage behaviors. 
 
     
     
         10 . The system of  claim 8  comprising the one or more processors, further configured to:
 generate the one or more recommendations, wherein at least one of the one or more recommendations comprises information that is associated with the interpretation of the pictorial information. 
 
     
     
         11 . The system of  claim 10  comprising the one or more processors, further configured to:
 generate the one or more recommendations, wherein the at least one of the one or more recommendations comprise the information that is associated with the interpretation of the pictorial information, wherein the information comprises a reference to a physical object. 
 
     
     
         12 . The system of  claim 8  comprising the one or more processors, further configured to:
 generate the one or more recommendations, wherein at least one of the one or more recommendations is applied to modify an element of the computer-implemented system. 
 
     
     
         13 . The system of  claim 8  comprising the one or more processors, further configured to:
 generate the one or more recommendations, wherein the one or more recommendations are generated in response to an inferred focus of attention of the user with respect to the video content. 
 
     
     
         14 . The system of  claim 8  comprising the one or more processors, further configured to:
 generate a natural language-based explanation for the delivery of at least one of the one or more recommendations, wherein the explanation comprises reasoning that is associated with the generation of the recommendation; and 
 deliver the explanation to the user. 
 
     
     
         15 . A computer-implemented system comprising one or more processors configured to:
 access a plurality of usage behaviors associated with one or more users of a computer-implemented system;   infer a preference of a user of the computer-implemented system from the plurality of usage behaviors;   access video content, wherein the video content comprises pictorial-based information;   interpret the pictorial-based information by application of a computer-implemented neural network;   generate a recommendation in response to an oral communication by the user, wherein the recommendation is generated based upon the inferring of the preference and the interpreting of the pictorial information; and   deliver the recommendation to the user.   
     
     
         16 . The system of  claim 15  comprising the one or more processors, further configured to:
 infer the preference of the user of the computer-implemented system from the plurality of usage behaviors, wherein the inference of the preference is based upon the application of a plurality of inference weightings that are determined in accordance with usage behavior priority rules that are applied to the plurality of usage behaviors. 
 
     
     
         17 . The system of  claim 15  comprising the one or more processors, further configured to:
 generate the recommendation, wherein the recommendation comprises information that is associated with the interpretation of the pictorial information, wherein the information comprises a reference to a physical object. 
 
     
     
         18 . The system of  claim 15  comprising the one or more processors, further configured to:
 generate the recommendation, wherein the recommendation comprises natural language-based information that is in an audio format. 
 
     
     
         19 . The system of  claim 18  comprising the one or more processors, further configured to:
 generate a natural language-based explanation for the delivery of the recommendation, wherein the explanation comprises reasoning that is associated with the generating of the recommendation; and 
 deliver the explanation in an audio format to the user. 
 
     
     
         20 . The system of  claim 19  comprising the one or more processors, further configured to:
 generate the natural language-based explanation, wherein the natural language language-based explanation further comprises a reference to at least one of the plurality of usage behaviors.