GB2487181A

GB2487181A - System for user profiling

Info

Publication number: GB2487181A
Application number: GB1022056.4A
Authority: GB
Inventors: Martin Hennelly
Original assignee: NEC Casio Mobile Communications Ltd
Current assignee: NEC Casio Mobile Communications Ltd
Priority date: 2010-12-29
Filing date: 2010-12-29
Publication date: 2012-07-18
Also published as: GB201022056D0

Abstract

A computer system is provided comprising a user device associated with a user and a profiling server. The user device 1 has a plurality of installed software applications 11 and a profiling agent that transmits data identifying the applications installed on the user device to the profiling server 21. The profiling server 21 has a receiver that receives data identifying applications installed on the user device and a profile generator that processes the received data to generate user profile data for the user associated with the user device in dependence upon the applications installed on the user device. The user profile data can then be used by service provider 25 to identify users to target with a new product or service advertisement.

Description

Computer System The present invention relates to computer systems and in particular to systems and methods for user profiling to facilitate the targeted delivery of products and services.

User profiling is well known. The aim of user profiling is to categorise a user into one or more categories in order to target products and services that may be of interest to users in that category. For example if it is known that a particular user likes football, then the targeted advertising of football related products and services to that user is more likely to be successful than traditional non-targeted advertising.

This invention relates to ways of obtaining user profile information that can be used in such targeted advertising systems.

According to one aspect, the present invention provides a computer system comprising: a user device (such as a portable computer device) associated with a user, the user device comprising: a plurality of installed software applications; a common repository for storing apphcation data relating to the installed software applications, wherein each software application is configured to write application data into the repository; and a profiling agent configured to process the application data stored in the repository and to transmit data identifying the applications installed on the user device; and a profiling server comprising: a receiver configured to receive data identifying applications installed on said user device; and a profile generator configured to process the received data and to generate user profile data for the user associated with the user device in dependence upon the applications installed on the user device.

The user device may be configured to maintain application usage data indicative of how each installed application is used by the associated user. In this case, the profiling agent may be configured to transmit data indicative of the application usage data for one or more installed applications to the profiling server and the profile generator may be configured to generate the user profile data in dependence upon the received application usage data.

In one embodiment, the application usage data for an application includes one or more of: the frequency with which the associated user uses the application, the last time that the application was used, the time of day when the application is used, where the user is when the application is used, how long the application is used for each time it its used, the volume of input/key-press/data the application receives, the difference in time between when the application was last used and the time when the application was installed or downloaded, the quantity of metadata the application generates or the portion of time for which user input into the application is active. This application usage data may be stored within the common repository or separately from said common repository. The application data maintained in the common repository may also include data identifying if the user has paid for an installed application.

The profiling agent may process the application data and determine a relative weighting of each installed application to the user and provide this determined weighting for each installed application to the profiling server In one embodiment, the profiling server determines one or more search tags associated with the installed applications which are included in the user profile data. The profiling server may also receive, from the user device or from a network device associated with the user device, location data for the user device and may store the location data as part of the user profile.

The present invention also provides a user device associated with a user, the user device comprising: a plurality of installed software applications; a common repository for storing application data relating to the installed software applications, wherein each software application is configured to write application data into the repository; and a profiling agent configured to process the application data stored in the repository and to transmit data identifying the applications installed on the user device.

The user device may maintain application usage data indicative of how each installed application is used by the associated user and the profiling agent may transmit data indicative of the application usage data for one or more installed applications to a profiling server.

The application usage data for an application may include one or more of: the frequency with which the associated user uses the application, the last time that the application was used, the time of day when the application is used, where the user is when the application is used, how long the application is used for each time it its used, the volume of input/key-press/data the application receives, the difference in time between when the application was last used and the time when the application was installed or downloaded, the quantity of metaclata the application generates or the portion of time for which user input into the application is active.

The application usage data may be stored within the common repository or separately from the common repository.

The application data may include data identifying if the user has paid for an installed application.

The profiling agent may process the application data and determine a relative weighting of each installed application to the user and to provide this determined weighting for each installed application to the profiling server.

The present invention also provides a profiling server comprising: a receiver configured to receive data identifying applications installed on a user device; and a profile generator configured to process the received data and to generate user profile data for a user associated with the user device in dependence upon the applications installed on the user device.

The profiling server may determine one or more search tags associated with the installed applications that are included within the user profile data.

The server may also receive, from the user device or from a network device associated with the user device, location data for the user device and may store location data as part of the user profile.

The profile generator may categorise the user into one or more profile categories in dependence upon the applications installed in the user device and the user profile may comprises data identifying the categorisation.

The present invention also provides a query server for use in identifying users to target1 the server comprising: an input for receiving a request for identifying users having a particular interest or characteristic; a query generator configured: i) to process the request and to generate a query relating to applications that may be installed in user devices; ii) to provide the query to a user profile database containing user profiles for a plurality of different users, each user profile comprising data identifying software applications installed on a user device associated with the user; and iii) to identify users corresponding to the request in dependence upon query results obtained from said database.

The query generator may process the request to identify applications that may be installed on user devices that are indicative of users having the particular interest or characteristic and may generate the query in dependence upon the identified applications.

The present invention also provides a user profile database comprising user profile data for each of a plurality of users, wherein each user profile comprises data identifying software applications installed on a user device associated with the user.

According to another aspect, the invention provides a method of user profiling characterised by the step of generating user profile data for a user in dependence upon software applications installed on a user device associated with the user.

The invention also provides a computer implementablo instructions product comprising computer implementable instructions for causing a computer device to be configured as the above user device, or to become configured as the above profiling server, or to become configured as the above query server, or to pertorm the above user profiling method.

These and other features and aspects of the invention will become apparent from the following detailed description of an exemplary embodiment described with reference to the accompanying drawings in which: Figure 1 is a block diagram illustrating the main components of a computer system according to an embodiment of the present invention; Figure 2 illustrates in more detail the software components residing on a user's device forming part of the system shown in Figure 1; Figure 3 is a block diagram illustrating the main components of a profiling server shown in Figure 1; Figure 4 is a block diagram illustrating the main components of a service provider server forming part of the system shown in Figure 1; and Figure 5 illustrates the way in which the user profile data can be searched to identify users to target for product/service or advertising delivery.

Overview In the past, software applications residing on a user's computer (laptop, desktop, portable etc) were generic (e.g. word processor, spreadsheet, browser etc) and data is often stored in user generated files. Recently, software applications have become more specific for a narrow domain or task. For example, there are now applications that count calories, provide the weather in a specific locale, provide traffic on a specific route and user data is gathered, presented and stored by the application itself rather than the user directly.

Software applications are also becoming more specific to tailored service delivery and are also downloaded into the user's computer device as a result of their "consumer choice". Thus the applications installed on a computing device are quite revealing about the interests and the preferences of the consumer. in the embodiment described below, a profiling sewer determines the software applications that are on the user's computing device, how they are used and the importance of the application to the user and uses this information to understand the characteristics of the user and their interests, preferences, habits and personality. The information obtained can then subsequently be used to deilver other services, e.g. services targeted to the user.

Figure 1 is a black diagram illustrating the main components of a system that allows user profile data to be determined and used subsequently to control the delivery of these other services. As shown, the system includes a user computer device 1 that is associated with a user. Multiple such user computer devices 1 will be provided for different users, but are not shown for clarity. The user device 1 may be a laptop, a desktop, a Personal Digital Assistant (PDA) or a mobile/cellular telephone. As shown, the user device 1 comprises a processor 3 which operates in accordance with software modules stored in memory 5. The software modules include an operating system 7, a web browser 9 and a number of applications 11-1 to 11-3. The user device I also includes a communications module 13, which allows the user device 1 to be able to connect to remote computer devices over a computer or telephone network 14. The user device 1 also includes a display 15 for outputting information to the user; and an input device 17, such as a keyboard and/or touchpad and/or mouse, for receiving user inputs to control the operation of the user device I and to interact with running applications.

Figure 1 also shows a profiling sewer 21 that interacts with the user's computer device 1 to determine what applications 11 are loaded on the user's computer device 1 and how the user is using them. From this information, the profiling server 21 generates user profile data that it stores in a user profile database 23.

This stored user profile data can then be used by a service provider server 25 to identify users to which a particular service or product should be targeted.

As will be described in more detail below, software applications 11, when installed on a user's computer device 1, write general metadata to a central (common) area in the computing system, such as a Windows or Linux registry or to a centrally managed plain text tile. This general metadata is stored under an identifier (e.g. the name of the application as a ext string) within the registry. The purpose of this is to store configuration data, preference data and to share parameters with other applications. The computer system 1 may also profile and store information on the application itself and this may be represented by an application usage protile.

Additionally, when the application is in use, events occur such as user input or storage to a database that indicates how the application is being used. This information has typically been used by, for example, a task manager to prompt the user to "remove unused icons from the desktop" to prevent the desktop from becoming too cluttered. In this embodiment, the information stored in the registry and any usage data for the installed applications is used to generate user profile data for the user associated with the user device 1, which user profile data can be used to control subsequent delivery of products or services targeted to the user.

The way in which this is achieved in an exemplary embodiment will now be described.

User's Computer Device Figure 2 illustrates in more detail the software modules and data stored within the user's computer device 1 in this embodiment. As shown, three applications 11-1, 11-2 and 11-3 (also labelled "A", "B" and "C") are loaded into the user's device 1.

Figure 2 also shows the usual application and system frameworks 31 forming part of the operating system 7, which comprises an application launcher 41 for launching applications; an event manager 43 for managing events with launched applications; an applications manager 45 tor managing the application lifecycle; and an inter-task message 46 via which applications 11 can communicate and share parameters. The application and system frameworks 31 also include the above mentioned registry 47; a File System (FS) 49 and a database 51 in which application data, including usage profile, may be stored for each of the installed applications 11.

The other frameworks illustrated in Figure 2 include a UI Framework 61 having a display module 63 for controlling the display 15 and a user input module 65 for accepting and interpreting user inputs received via the user input 17; and ancillary frameworks 67 such as a payment framework 69 for controlling the making and receiving of payments and a location framework 71 for determining the location of the user's computer device etc..

The memory 5 may also include a web runtime module 73 that allows standard applications from the web to be run on the platform.

Finally, as shown in Figure 2, the memory 5 also includes an application profiling agent 75 that processes the registry file 47 and any application usage profiles to analyse applications 11 installed on the computing device 1, and that assembles an application profile in co-operation with other system frameworks and enablers, e.g. the application manager 45, user input framework 65, payment framework 69 etc. In this embodiment, the application profiling agent 75 processes the gathered information to determine the relative weighting and importance of each application to the user, so that, for example: a paid-for application is given a higher weighting than a free application.

For example, the profiling agent 75 can interact with the payment framework 69, if it exists, at the time of purchase to determine which applications are paid. Or, if applications were paid for directly within a browser session, the profiling agent 75 can monitor the browser session to determine payment events and the download of a new application or by analysis of the content of the application download site.

* applications used more frequently are given a higher weighting than those used less often * the weighting given to an application depends on when the application was last used * the weighting given to an application depends on the time of day when the application is used * the weighting given to an application depends on where the user is when the application is used * the weighting given to an application depends on how long the application is used for each time it its used * the weighting given to an application that was last used shortly after the time at which it was download is given a low weighting * the weighting given to an application depends on the quantity of other metadata the application generates * the weighting given to an application depends on its relevance to indicating user characteristics -e.g. the browser 9 is generally a high use application, but by itself reveals litile about the user: although metadata such as cookies and history reveals a lot * Portion of time for which input is active, e.g. a messaging service application may be always executing, but the portion of active user time may be low Pro fifing Server Thus the profiling agent 75 collects the data on the user device on each individual application and this data is supplied to the profiling server 21, which processes the data and turns it into an overall profile of the user by considering the whole set of applications, which the profile server 21 then makes available for other services.

Typically, the profile server 21 will have available metadata which is a catalogue of application name (or other identifier) and application function. This catalogue can be compiled from a web search of application stores, analysing the application stores own category and then further analysis of the description. Further refinement of the category could be made by human analysis. Once the profiling server 21 has categorised the application, the application can then be tagged with relevant searchable terms through the application of a rules based system. The total analysis of the applications (of the categories and searchable terms) may then be aggregated according to the weighting of the importance of the application to the user as determined by the statistical analysis by the profiling agent 75. The profiling server 21 may also determine further weightings to be applied to the applications. For example, the profiling server 21 may analyse the function of the application and the amount of user input to determine a measure of the importance of the application to the user and then weight the applications so that important applications are given more weighting than less important applications.

The profiling server 21 can also tag the data with the typical location at the user.

The resulting user profile data generated by the profiling server 21 is then stored in the user profile database 23 for subsequent use by a service provider server 25 in delivering relevant and applicable products and/or services orientated to users via their computing devices 1.

Figure 3 is a block diagram illustrating the main components of the profiling server 21 used in this embodiment. As shown, the profiling sewer 21 comprises a processor 81 which operates in accordance with software modules stored in memory 83. The software modules include an operating system 85, a profile generator 87 and a user data store 89. The profiling server 21 also includes a communications module 91, which allows the profiling server 21 to be able to connect to remote user devices 1 over the network 14 in order to obtain user profile data from the application profiling agent 75; and to be able to search for additional application data from other servers (not shown) coupled to the network 14 to add to the user profile data. The profile server 21 also includes a user interface 93 for providing input/output to an administrator of the profile server 21; and a user profile database interface 95 for interlacing with the user profile database 23.

In operation, the profile generator 87 would receive user data from the application profiling agent 75 in a user computer device 1. The profile generator may poll the user device 1 for this information or the application profiling agent 75 may send the data from time to time or when a new application is installed onto the user's computer device 1. The received user data is stored in the user data store 89 and processed by the profile generator. As discussed above, the profile generator 87 processes the user data to build a searchable dataset of tags that describe the user's interests (derived from analysis of the applications downloaded and their usage patterns) as well as their typical locations. Once the user profile data has been processed and tagged, the profile generator 87 stores the user profile data in the user profile database 23 in association with the user, for subsequent use in targeting relevant advertisements or services to that user.

Service Provider Server Figure 4 is a block diagram illustrating the main components of the service provider server 25 used in this embodiment. As shown, the service provider server 25 comprises a processor 101 which operates in accordance with software modules stored in memory 103. The software modules include an operating system 105 and a query generator 107. The service provider server 25 also includes a communications module 111, which allows the service provider sewer to be able to connect to remote user devices 1 over the network 14 in order to provide targeted services or advertisements to selected user computer devices 1.

The service provider server 25 also includes a user interlace 113 for providing input/output to an administrator of the service provider server 25; and a user profile database interface 115 for interfacing with the user profile database 23.

In operation, an administrator of the service provider server inputs the requirements for a search of the user profile database 23 via the user interface 113. Alternatively, the search requirements may be received via the communications module 111 from another computer device coupled to the network 14. In response to receiving the search requirements, the query generator processes the search requirements to generate appropriate search terms for application to the user profile database 23. The search terms may be the names of specific applications installed on user devices or location information or generalised search terms that correspond to one or more of the tags added by the profiling server 21. The determined search terms are then applied to the user profile database 23 via the interface 115. The results of the query are then returned to the service provider server 25 which uses the results to identify different users meeting the search requirements and for whom the service or advertisement is to be targeted.

The process is illustrated in Figure 5, which shows part of the user profile database 23. As shown, the database 23 includes a number of different user profiles 121, each being associated with a user location and user applications installed on their device. In this example, the search terms identify a location and two applications (A&B). The profiles in the database 23 are then ordered based on their associated locations and installed applications, so that the user profiles matching the search terms can be quickly identified and data identifying the relevant users is returned to the service provider server 25.

As an example of the benefit of the system described above, consider the case where the profiling agent 75 on the user's computer device 1 provides location information together with the application profile data. Location, with varying degrees of accuracy, is a freely available parameter in computing devices. E.g. in a desktop computer it may be inferred from the ISP (Internet Service Provider's) parameters; in a mobile device it may be available from (e.g.) GPS, mobile phone system cell ID, WiFi parameters etc. For exampre, a service provider server 25 with access to the user profile information and location information in the database 23 may use this information to compile a data set for an analysis of the preferences and interests of the consumers in different geographical regions. The result of that analysis may then be used to deliver a targeted service.

The analysis performed by the service provider server 25 could be performed using specific search terms, e.g. England Cricket fans in Birmingham, or as a general profiling of the users in the data set by their location. Potential purposes of such analysis would be for building communities of people with common interests in a geographical area and for targeted advertising services, particularly for local services.

The operation of the above embodiment may be summarised as follows: -Respective application metadata is collected by a number of user devices 1 and sent to a central profiling server 21 (or a local profiling server 21 in the case of a distributed architecture).

-Further analysis of the application metadata is performed at the profiling server 21, e.g. tagging with search terms and if available with location information (e.g. home location, office location, current location, etc.) obtained from the user device or from the computer network 14.

-Analysis of Application Metadata with location across multiple users in the system on the service provider server 25 based on search terms' originating from a service requestor.

-Offer service to users based on the above results.

-Or by overall profiling of users in the system by location.

Example scenarios

A running community could be created in a local area. Users with the native application Runner' (sportstracklive.com), F{unstar' or Nokia sports tracker' can be searched out in a local area and a service offering made. For example, a user may wish to find other runners in their local area and could seek a 1'running buddy" to the service provider server 25. In response, the service provider sewer 25 would search the user profile database 23 for other users nearby the requesting user who may also be interested in running (determined from the presence of one or more of the above applications on the users computing device); and would then broker an introduction between the requesting user and the users found in the search. Alternatively, a sport equipment manufacturer or a sports club may be interested in identifying runners in a particular local area for a special promotion that they are having. In this case the service provider server 25 would identify the relevant users and would send those users details of the special promotion being offered by the requester.

Smoking cessation support partners could be found in a particular local area by finding other people with a quit counter' application. In this case, the service provider server 25 could broker a meeting or other support contact with the requesting user.

Similarly, a dieting support partner could be found by finding users with a calorie counter' or other dieting application installed on their devices.

Shared rides to football matches could be brokered to fans with a widget (installed application) that tracks their regional football club and living in the same area.

A targeted local advertising service could also be delivered based on the user profile data, e.g. a bookshop in Reading may be running a book signing by a celebrity chef and wants to target people who would be interested. Users with devices having a recipe widget' or a restaurant review widgef within 20 mIles of Reading could then be found by the service provider server 25 using the data in the user profile database 23 and then targeted to receive the advertising.

Modifications and Alternatives A detailed embodiment has been described above. As those skilled in the art will appreciate, a number of modifications and alternatives can be made to the above embodiment whilst still benefiting from the invention embodied therein.

In the above embodiment, a central profiling server 21 was used to process application metadata received from a user device, typically adding search tags to the data and storing the data in the user profile database 23. The processing pertormed by the profiling server 21 could all be performed by the profiling agent loaded in the user's computer device 1. Alternatively, the profiling agent 75 may perform less analysis than mentioned above, instead just gathering the application metadata for onward transmission to the profiling server 21.

In the above embodiments, the information about what applications are stored on the user's computing device came from a profiling agent stored on the user's computing device. In addition or alternatively, this application metadata may be provided by other web servers. For example, when a user buys an application from a website, the details of the user and the application may be passed to the profiling server 21 for use in generating the relevant user profile data in the user profile database 23.

In the above embodiment, one profiling server 21 and one service provider server 25 were described. In a distributed system, multiple profiling servers and service provider servers may be provided in different geographical regions.

In the above embodiment, the profiling server 21 was separate from the service provider server 25. In an alternative embodiment, the functionality of these two servers may be performed by a single server.

The various modules described above are typically software modules. They may be separate modules or the functionality of one or more of the modules may be performed by a single module or by a number of separate modules. The software may be provided on a carrier medium, such as a CD or DVD and may be downloaded over a computer network into the user's device or server.

These and various other modifications and alternatives will be apparent to those skilled in the art.

Claims

Claims 1. A computer system comprising: a user device associated with a user, the user device comprising: a plurality of installed software applications; a common repository for storing application data relating to the installed software applications, wherein each software application is configured to write application data into the repository; and a profiling agent configured to process the application data stored in the repository and to transmit data identifying the applications installed on the user device; and a profiling server comprising: a receiver configured to receive data identifying applications installed on said user device; and a profile generator configured to process the received data and to generate user profile data for the user associated with the user device in dependence upon the applications installed on the user device.
2. A system according to claim 1, wherein the user device is configured to maintain application usage data indicative of how each installed application is used by the associated user, wherein said profiling agent is configured to transmit data indicative of the application usage data for one or more installed applications to the profiling server and wherein the profile generator is configured to generate the user profile data in dependence upon the received application usage data.
3. A system according to claim 2, wherein the application usage data for an application includes one or more of: the frequency with which the associated user uses the application, the last time that the application was used, the time of day when the application is used, where the user is when the application is used, how long the application is used for each time it its used, the volume of input/key-press/data the application receives, the difference in time between when the application was last used and the time when the application was installed or downloaded, the quantity of metadata the application generates or the portion of time for which user input into the application is active.
4. A system according to claim 3, wherein said application usage data is stored within said common repository or separately from said common repository.
5. A system according to any of claims 1 to 4, wherein the application data maintained in said common repository includes data identifying if the user has paid for an installed application.
6. A system according to any of claims 1 to 5, wherein said profiling agent is configured to process the application data and to determine a relative weighting of each installed application to the user and to provide this determined weighting for each installed application to the profiling server.
7. A system according to any of claims 1 to 6, wherein the profiling server is configured to determine one or more search tags associated with the installed applications and wherein the user profile data includes the determined search tags.
8 A system according to any of claims 1 to 7, wherein the profiling server is configured to receive, from the user device or from a network device associated with the user device, location data for the user device and is configured to store location data as part of the user profile.
9. A user device associated with a user, the user device comprising: a plurality of installed software applications; a common repository for storing application data relating to the installed software applications, wherein each software application is configured to write application data into the repository; and a profiling agent configured to process the application data stored in the repository and to transmit data identifying the applications instalred on the user device.
10. A device according to claim 9, configured to maintain application usage data indicative of how each installed application is used by the associated user and wherein said profiling agent is configured to transmit data indicative of the application usage data for one or more installed applications to a profiling server.
11. A device according to claim 10, wherein the application usage data for an application includes one or more of: the frequency with which the associated user uses the application, the last time that the application was used, the time of day when the application is used, where the user is when the application is used, how long the application is used for each time it its used, the volume of input/key-press/data the application receives, the difference in time between when the application was last used and the time when the application was installed or downloaded, the quantity of metadata the application generates or the portion of time (or which user input into the application is active.
12. A device according to claim 11, wherein said application usage data is stored within said common repository or separately from said common repository.
13. A device according to any of claims 9 to 12, wherein the application data maintained in said common repository includes data identifying if the user has paid for an installed application.
14. A device according to any of claims 9 to 13, wherein said profiling agent is configured to process the application data and to determine a relative weighting of each installed application to the user and to provide this determined weighting for each installed application to the profiling server.
15. A profiling server comprising: a receiver configured to receive data identifying applications installed on a user device; and a profile generator configured to process the received data and to generate user profile data for a user associated with the user device in dependence upon the applications installed on the user device.
16. A server according to claim 15, configured to determine one or more search tags associated with the installed applications and wherein the user profile data includes the determined search tags.
17. A server according to claim 15 or 16, configured to receive, from the user device or from a network device associated with the user device, location data for the user device and configured to store location data as part of the user profile.
18. A server according to any of claims 15 to 17, wherein said profile generator is configured to categorise the user into one or more profile categories in dependence upon the applications installed in the user device and wherein the user profile comprises data identifying said categorisation.
19. A server for use in identifying users to target, the server comprising: an input for receiving a request for identifying users having a particular interest or characteristic; a query generator configured: i) to process the request and to generate a query relating to applications that may be installed in user devices; ii) to provide the query to a user profile database containing user profiles for a plurality of different users, each user profile comprising data identifying software applications installed on a user device associated with the user; and iii) to identify users corresponding to the request in dependence upon query results obtained from said database.
20. A server according to claim 19, wherein the query generator is configured to process the request to identify applications that may be installed on user devices that are indicative of users having the particular interest or characteristic and is configured to generate the query in dependence upon the identified applications.
21. A user profile database comprising user profile data for each of a plurality of users, wherein each user profile comprises data identifying software applications installed on a user device associated with the user.
22. A method of user profiling characterised by the step of generating user profile data for a user in dependence upon software applications installed on a user device associated with the user.
23. A computer implementable instructions product comprising computer implementable instructions for causing a computer device to be configured as the user device of any of claims 9 to 14, or to become configured as the profiling server of any of claims 15 to 18, or to become configured as the server of claim 19 or 20, or to perform the method of claim 22.