Recommending mobile device activities

Abstract

Techniques for recommending mobile device activities, such as accessing mobile applications and/or mobile Web pages, are described. Some embodiments provide an Activity Recommendation System (“ARS”) configured to recommend relevant activities for a user to perform with a mobile device, based on context of the mobile device. In one embodiment, the ARS recommends mobile applications based content items (e.g., Web pages, images, videos) that are being currently accessed via the mobile device. The ARS may process information about mobile applications and content items to determine semantic information, such as entities and/or categories referenced or associated therewith. The ARS may then use the semantic information to determine mobile applications that have semantic information that is at least similar to that of a content item accessed via a mobile device.

1 - 31 . (canceled) 
     
     
         32 . A computer-implemented method for discovering mobile applications, the method comprising:
 generating in a data store of a computing system an index of semantic information for multiple mobile applications from a plurality of sources including at least one of a Web site, a blog, a news feed, a video feed, a social network, an application store, and/or an information service, the semantic information including identified entities and/or categories related to the multiple mobile applications, wherein the identification is performed using natural language processing techniques;   receiving a semantic search query, the query specifying at least one entity or category;   determining from the index one or more of the multiple mobile applications that have associated semantic information matching the received semantic search query or that have associated semantic information that is related to the received semantic search query through an entity and/or category that is related to an entity and/or category stored in the index of semantic information; and   transmitting indications of the determined one or more mobile applications in response to the semantic search query.   
     
     
         33 . The method of  claim 32  wherein generating the index includes processing information about each of the multiple mobile applications, including at least one of a summary of the application, a description of the application, a title for the application, instructions for using the application, and a review of the application. 
     
     
         34 . The method of  claim 33  wherein processing the information about each of the multiple mobile applications includes determining an entity and/or a category referenced by the processed information and storing a relationship between the entity and/or category referenced and a corresponding entity and/or category stored in the data store. 
     
     
         35 . The method of  claim 32  wherein generating the index includes storing a mapping between an entity or category and one or more of the multiple mobile applications that are related to the entity or category. 
     
     
         36 . The method of  claim 32  wherein determining from the index the one or more mobile applications that have associated semantic information matching the received semantic search query includes accessing the index to locate mobile applications that are mapped to the entity or category specified by the semantic search query. 
     
     
         37 . The method of  claim 32  wherein receiving the semantic search query includes receiving a search query that requests mobile applications about a specified entity or category, and wherein determining from the index the one or more mobile applications includes locating one or more mobile applications that are about the specified entity or category. 
     
     
         38 . The method of  claim 32  wherein receiving the semantic search query includes receiving a search query that requests mobile applications that are indirectly related to a specified entity or category, and wherein determining from the index the one or more mobile applications includes locating one or more mobile applications that are about an entity or category that is related to the specified entity or category. 
     
     
         39 . The method of  claim 32  wherein the multiple mobile applications include applications that execute at least in part on mobile devices. 
     
     
         40 . The method of  claim 32  wherein the multiple mobile applications include Web sites configured for use by mobile devices. 
     
     
         41 . The method of  claim 32  wherein receiving the semantic search query includes receiving the semantic search query from a mobile application recommender module executing on a mobile device, the recommender module configured to transmit the semantic search query in response to receiving indication of access to a content item accessed via the mobile device, the content item referencing the specified at least one entity or category. 
     
     
         42 . A computing system configured to discover mobile applications comprising:
 a computer processor;   a memory;   code logic that is stored in the memory and that is configured, when executed by the computer processor, to:
 generate an index in a data store in the memory of the computing system of semantic information for multiple mobile applications from a plurality of sources including at least one of a Web site, a blog, a news feed, a video feed, a social network, an application store, and/or an information service, the semantic information including identified entities and/or categories related to the multiple mobile applications, wherein the identification is performed using natural language processing techniques; 
 receive a semantic search query, the query specifying at least one entity or category; 
 determine from the index one or more of the multiple mobile applications that have associated semantic information matching the received semantic search query or that have associated semantic information that is related to the received semantic search query through an entity and/or category that is related to an entity and/or category stored in the index of semantic information; and 
 transmit indications of the determined one or more mobile applications in response to the semantic search query. 
   
     
     
         43 . The computing system of  claim 42  wherein the code logic includes software instructions for execution in the memory of the computing system. 
     
     
         44 . The computing system of  claim 42  wherein the code logic includes an activity recommendation system. 
     
     
         45 . The computing system of  claim 44  wherein the activity recommendation system includes a client portion executing on a mobile device. 
     
     
         46 . The computing system of  claim 42  wherein the code logic is configured to recommend activities to at least one of a personal digital assistant, a smart phone, a laptop computer, a notebook computer, a mobile phone, and/or a third-party application. 
     
     
         47 . The computing system of  claim 42  wherein the mobile device is a tablet computer. 
     
     
         48 . The computing system of  claim 42  wherein the data store includes an application index and an entity store. 
     
     
         49 . The computing system of  claim 42  wherein the code logic generates the index about each of the multiple mobile applications by determining an entity and/or a category referenced by the semantic information and storing a relationship between the to entity and/or category referenced and the corresponding entity and/or category in the data store. 
     
     
         50 . A non-transitory computer-readable medium containing instructions for controlling a computer processor to discover mobile applications by performing method comprising:
 generating in a data store of a computing system an index of semantic information for multiple mobile applications from a plurality of sources including at least one of a Web site, a blog, a news feed, a video feed, a social network, an application store, and/or an information service, the semantic information including identified entities and/or categories related to the multiple mobile applications, wherein the identification is performed using natural language processing techniques;   receiving a semantic search query, the query specifying at least one entity or category;   determining from the index one or more of the multiple mobile applications that have associated semantic information matching the received semantic search query or that have associated semantic information that is related to the received semantic search query through an entity and/or category that is related to an entity and/or category stored in the index of semantic information; and   transmitting indications of the determined one or more mobile applications in response to the semantic search query.   
     
     
         51 . The computer-readable medium of  claim 50  wherein the instructions are code logic, a portion of which is stored and executes on a mobile device.