Virtual Tiles For Service Content Recommendation

Patent metadata

Field	Value
Publication number	US-2017031996-A1
Application number	US-201514811289-A
Country	US
Kind code	A1
Filing date	Jul 28, 2015
Priority date	Jul 28, 2015
Publication date	Feb 2, 2017
Grant date	—

Abstract

In various implementations, virtual tiles are constructed where each virtual tile includes geographic tiles having one or more shared semantic characteristics. Usage records of service content items are received, which are generated from sensor data from a plurality of mobile devices within a geographic region that includes the geographic tiles. Each usage record identifies a service content item, and a usage location corresponding to a user interaction with the service content item. Statistical correlations are calculated between service content items and a virtual tile based on an aggregation of the usage records that have a usage location within at least one of the geographic tiles of the virtual tile. Recommendations of one of more service content items are provided to a user based on the calculated statistical correlations. In some cases, one or more service content items are recommended based on being representative of a particular virtual tile.

First claim

What is claimed is:

1 . A computerized method comprising:
identifying a location of a user based on sensor data provided by one or more sensors of a user device; determining one or more semantic characteristics of the location of the user; matching the one or more semantic characteristics of the location of the user to one or more semantic characteristics of a virtual tile comprising geographic tiles, each geographic tile corresponding to a designated geographic area, the one or more shared semantic characteristics being of the geographic area of the geographic tile; receiving usage records of service content items, the usage records generated from sensor data provided by one or more sensors of a plurality of mobile devices within a geographic region that includes the geographic tiles, each usage record identifying a service content item of the service content items, and comprising a usage location corresponding to a user interaction with the service content item; calculating a statistical correlation between at least one of the service content items and the virtual tile based on an aggregation of the usage records, the aggregation being based on identifying ones usage records that have the usage location of the at least one of the service content items within the geographic tiles of the virtual tile; and providing a recommendation of one of more service content items to the user based on the calculated statistical correlation.

2 . The method of claim 1 , wherein the identifying the location of the user comprises identifying a venue visited by the user and the one or more semantic characteristics correspond to a category of the venue.

3 . The method of claim 1 , wherein the one or more shared semantic characteristics correspond to at least one venue of a category of venue being located within the designated geographic area for each of the geographic tiles.

4 . The method of claim 1 , wherein each of the service content items is an app and the user interaction corresponds to a launch of the app.

5 . The method of claim 1 , wherein the calculated statistical correlation is between usage of the at least one of the service content items and the virtual tile.

6 . The method of claim 1 , wherein the calculated statistical correlation represents a probability the aggregation of the usage records were observed in the virtual tile by chance.

7 . The method of claim 1 , wherein the matching is further of one or more semantic characteristics of the user to the one or more semantic characteristics of the virtual tile.

8 . The method of claim 1 , wherein the location is outside of the virtual tile.

9 . The method of claim 1 , wherein the matching the one or more semantic characteristics of the location of the user to one or more semantic characteristics of the virtual tile comprises identifying a plurality of pre-constructed geographic tiles for the virtual tile based on the pre-constructed geographic tiles having semantic characteristic values corresponding to the one or more semantic characteristics of the location.

10 . A computerized system comprising:
a content recommendation engine comprising one or more processors; and one or more computer storage media storing computer-useable instructions that, when executed by the one or more processors, cause the one or more processors to perform operations comprising:
constructing a virtual tile comprising geographic tiles based on identifying one or more shared semantic characteristics amongst the geographic tiles, each geographic tile corresponding to a designated geographic area, the one or more shared semantic characteristics being of the geographic area of the geographic tile;
receiving usage records of service content items, the usage records generated from sensor data provided by one or more sensors of a plurality of mobile devices within a geographic region that includes the geographic tiles, each usage record identifying a service content item of the service content items, and comprising a usage location corresponding to a user interaction with the service content item;
calculating a statistical correlation between at least one of the service content items and the virtual tile based on an aggregation of the usage records, the aggregation being based on identifying ones of the usage records that have the usage location of the at least one of the service content items within the geographic tiles of the virtual tile; and
providing a recommendation of one of more service content items to a user based on the calculated statistical correlation.

11 . The system of claim 10 , wherein the constructing the virtual tile comprises:
identifying venue profiles corresponding to venues of a particular category of venue, the venue profiles comprising locations of the venues; generating one of the geographic tiles of the virtual tile for each of the venues based on the locations of the venues, wherein the particular category of venue is one of the one or more shared semantic characteristics.

12 . The system of claim 10 , wherein the one or more shared semantic characteristics correspond to at least one venue of a particular venue chain being located within the designated geographic area for each of the geographic tiles.

13 . The system of claim 10 , comprising:
determining that a venue visited by the user is in a particular category of venue; including the at least one of the service content items in the recommendation provided to the user based on the particular category of venue matching the one or more shared semantic characteristics amongst the geographic tiles and based on the calculated statistical correlation.

14 . The system of claim 10 , comprising:
tracking instances of one or more events of the user, the events being of a routine tracked by the system; determining that the user practices the routine based on the tracked instances of the one or more events; inferring a routine characteristic of the user based on the determining that the user practices the routine; including the at least one of the service content items in the recommendation provided to the user based on the routine characteristic of the user matching the one or more shared semantic characteristics amongst the geographic tiles and based on the calculated statistical correlation.

15 . The system of claim 10 , wherein the calculated statistical correlation represents a probability the aggregation of the usage records were observed in the virtual tile by chance.

16 . The system of claim 10 , wherein the calculated semantic characteristic is one of a plurality of features of a classifier trained to identify service content items that are representative of the virtual tile.

17 . The system of claim 10 , wherein the providing the recommendation of the one of more service content items to the user is further based on a usage record associated with the user, the usage record corresponding to the at least one of the service content items.

18 . One or more computer storage devices storing computer-useable instructions that, when used by one or more computing devices, cause the one or more computing devices to perform a method comprising:
identifying an interaction between a user and a service content item; identifying a statistical correlation between the service content item and a virtual tile, the statistical correlation based on an aggregation of usage records, the aggregation being based on identifying ones of the usage records that have a usage location of the service content item within geographic tiles of t

Assignees

• Microsoft Technology Licensing Llc

Inventors

• Priness Ido
• Dotan-Cohen Dikla
• Sapir Boaz
• Avigdor Daniel

Classifications

• G06F16/9535

  Search customisation based on user profiles and personalisation · CPC title

• G06F16/248

  Presentation of query results · CPC title

• G06F16/9537

  Spatial or temporal dependent retrieval, e.g. spatiotemporal queries · CPC title

• G06F16/24575 Primary

  using context · CPC title

• G06Q30/06 Primary

  Buying, selling or leasing transactions · CPC title