Automatic Generation of Preferred Views for Personal Content Collections

Abstract

A computer system obtains access to a content collection that includes a plurality of content items. For each of the content items, the computer system determines user access history and context associated with the access history. The system builds a classifier characterizing a user access pattern of the content items. The system constantly monitors in real-time environmental parameters of the user. It infers a current context of the user based on the environmental parameters. In accordance with the current context, for each of the content items, the system extracts numeric feature values, evaluates the item using the classifier, and determines a score for the item. The system identifies in real-time a subset of the content items that is most relevant to the current context. The system further generates a preferred view of the subset and causes the preferred view to be delivered to a mobile device.

What is claimed is: 
     
         1 . A content management method, comprising:
 at a computer system including one or more processors and memory storing programs for execution by the one or more processors:
 obtaining access to a plurality of content items, each content item having a respective type, respective user access history, and respective historic context; 
 characterizing a user access pattern based on the respective user access history and historic context of each content item; and 
 while monitoring environmental parameters of a user, in real time:
 inferring a current context of the user based on the environmental parameters; 
 in accordance with the current context and the user access pattern, identifying a subset of content items having a highest likelihood of user access among the plurality of content items; and 
 causing to be delivered to, and displayed on, a mobile device a preferred view of the subset of content items. 
 
   
     
     
         2 . The method of  claim 1 , wherein the computer system is communicatively connected with a mobile device associated with the user, and the plurality of content items are obtained from a content collection of the user and have a plurality of content types. 
     
     
         3 . The method of  claim 1 , further comprising:
 for each of the plurality of content items, determining the respective user access history and the respective historic context;   building a classifier for characterizing the user access pattern; and   evaluating each content item using the classifier.   
     
     
         4 . The method of  claim 1 , in accordance with the current context and the user access pattern, identifying the subset of content items having the highest likelihood of user access among the plurality of content items further comprising:
 in accordance with the current context, for each of the plurality of content items:
 extracting numeric feature values to generate a vector of feature values; 
 determining a score for the respective content item based on the vector of feature values, the score indicating a likelihood of user access of the content item in the current context; and 
 identifying in real-time the subset of content items in the plurality of content items that is most relevant to the current context based on the determined score, wherein the subset of content items is a partial aggregation of the plurality of content items having highest scores. 
   
     
     
         5 . The method of  claim 1 , further comprising:
 causing to be delivered to, and displayed on, the mobile device a list of favorite user notes, wherein the preferred view populates a section of the list of favorite user notes, wherein the plurality of content items are displayed as a list on the mobile device, and the subset of content items is displayed on top of the list.   
     
     
         6 . The method of  claim 1 , wherein for each content item, the respective type is one of document, user note, notebook, search list, media file, appointment, navigational route, and reminder. 
     
     
         7 . The method of  claim 1 , wherein, for each content item:
 a context is associated with the respective user access history and includes a plurality of components; and   each of the components has a respective set of distinct features that is used to evaluate a relevance of the respective content item.   
     
     
         8 . The method of  claim 7 , wherein inferring the current context of the user further comprises comparing the respective set of features of each of the components in the context associated with the respective user access history and the current context. 
     
     
         9 . The method of  claim 7 , wherein the plurality of components includes a plurality of: temporal, geolocation, scheduled events, navigational, organizational, sharing context, search, travel, and social network. 
     
     
         10 . The method of  claim 1 , wherein for each content item:
 a context is associated with the respective user access history and includes a temporal context; and   the temporal context includes one or more temporal patterns of access of an item selected from: a time of recent access of an item, frequency of access of an item, frequency of location related access of an item, and frequency of event related access of an item.   
     
     
         11 . A computer system, comprising:
 one or more processors; and   memory storing one or more programs to be executed by the one or more processors, the one or more programs comprising instructions for:
 obtaining access to a plurality of content items, each content item having a respective type, respective user access history, and respective historic context; 
 characterizing a user access pattern based on the respective user access history and historic context of each content item; and 
 while monitoring environmental parameters of a user, in real time:
 inferring a current context of the user based on the environmental parameters; 
 in accordance with the current context and the user access pattern, identifying a subset of content items having a highest likelihood of user access among the plurality of content items; and 
 causing to be delivered to, and displayed on, a mobile device a preferred view of the subset of content items. 
 
   
     
     
         12 . The computer system of  claim 11 , wherein the one or more temporal patterns of access of the item are numerically assessed based on at least one of: time of day, time of week, and time of month. 
     
     
         13 . The computer system of  claim 11 , wherein:
 the subset of content items includes only content items having a relevance value above a predetermined threshold; and   the one or more programs further comprise instructions for:
 sorting the subset of content items according to respective relevance values of the content items in the subset of content items, and wherein content items that are not part of the subset of content items are displayed following the subset of content items. 
   
     
     
         14 . The computer system of  claim 11 , the one or more programs further comprising instructions for:
 splitting the content items into a training set and a test set; and   analyzing the content items in the training set to develop a set of rules used for evaluation of relevance of the content items; and   building a classifier for characterizing the user access pattern, wherein the classifier further includes instructions for performing automatic learning on the training set.   
     
     
         15 . The computer system of  claim 14 , wherein:
 the classifier comprises a support vector machine classification method; and   the one or more programs further comprising instructions for, for each content item:
 generating a vector of feature values; and 
 determining a score based on a relationship between the vector of feature values and a normal vector of a separating hyperplane using the support vector machine classification method. 
   
     
     
         16 . A non-transitory computer-readable storage medium storing one or more programs for execution by one or more processors of a computer system, the one or more programs comprising instructions for:
 obtaining access to a plurality of content items, each content item having a respective type, respective user access history, and respective historic context;   characterizing a user access pattern based on the respective user access history and historic context of each content item; and while monitoring environmental parameters of a user, in real time:
 inferring a current context of the user based on the environmental parameters; 
 in accordance with the current context and the user access pattern, identifying a subset of content items having a highest likelihood of user access among the plurality of content items; and 
 causing to be delivered to, and displayed on, a mobile device a preferred view of the subset of content items. 
   
     
     
         17 . The non-transitory computer-readable storage medium of  claim 16 , the one or more programs further comprising instructions for:
 adjusting subsequent subsets of the plurality of content items based on user feedback, wherein the user feedback is implicit and includes frequency of actual viewing of respective items of the plurality of content items by the user.   
     
     
         18 . The non-transitory computer-readable storage medium of  claim 16 , the one or more programs further comprising instructions for:
 adjusting subsequent subsets of the plurality of content items based on user feedback, wherein the user feedback is explicit.   
     
     
         19 . The non-transitory computer-readable storage medium of  claim 16 , wherein the preferred view is configured for presentation based on a ranking of each content item in the subset of content items. 
     
     
         20 . The non-transitory computer-readable storage medium of  claim 19 , wherein the preferred view includes a pop up screen that is superimposed over a first view of the plurality of content items.