WO2015186022A1

WO2015186022A1 - Method and system of identifying a target set of mobile device users

Info

Publication number: WO2015186022A1
Application number: PCT/IB2015/053960
Authority: WO
Inventors: Pranav KUMAR JHA; Ashwin RAMASWAMY; Raghvendra VARMA
Original assignee: Mubble Networks Private Limited
Priority date: 2014-06-04
Filing date: 2015-05-27
Publication date: 2015-12-10

Abstract

A method of identifying a target set of mobile device users is disclosed. The disclosed method comprises of collecting time series usage data for multiple usage parameters from mobile devices of a group of users, analyzing the usage data to identify clusters, such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on its priority rank. Further, the method comprises of analyzing each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster. Further, the method comprises of collecting user attitude data from each user of the group of users, identifying co-relations between the usage pattern and the user attitude data for a cluster and, on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.

Description

METHOD AND SYSTEM OF IDENTIFYING A TARGET SET OF MOBILE DEVICE USER

The present invention discloses a method and system of identifying a target set of mobile device users.

With the advent of smartphones in the market and subsequent increase in mobile device internet usage, marketing via mobile devices has become an integral part of marketing strategies for companies. Marketing via mobile devices benefits consumers, mobile service providers, publishers and marketers and drives higher revenues by targeting a specific set of users. This target set of mobile device users are identified to be the most likely consumers of the company’s services and products. Such target set of mobile device users are identified based on the analysis of their related mobile usage behaviour. The mobile usage behaviour is determined majorly from the volume of mobile data generated on the mobile devices. The mobile usage behaviour analytics is a powerful tool used by the marketers to find their target consumers. Although, various means of analysis of mobile usage data is known in the art, the publishers and marketers are continuously challenged to identify the most appropriate group of users i.e. the most likely users of the advertised service or product.

It is also known in the art to segment the mobile device users on the basis of the mobile data generated on the mobile devices. Many techniques such as data mining processes have been opted for segmenting the vast mobile usage data generated on the mobile devices. However, targeting the mobile device users only on the basis of the segmentation of the mobile data, which could be very large in volume in most cases, involves complex mechanisms and is not always reliable. Moreover, such systems inherently compromise the privacy of users. Further, the marketers face challenging tasks of relating the users to their actual preferences of services and products in reality.

It is also well known traditionally to collect user’s preferences of services and products manually through online market surveys or personal surveys in the form of attitude data or psychographic data of the users. Thereafter, such collected data is analyzed to segment or categorize the users on the basis of the general characteristics exhibited by the data collected from the users. Such traditional approach involves the manual collection, compilation and analysis of vast user data before segmenting the users on the basis of the general characteristics exhibited by a particular segment.

United States Patent Application Publication Number 2013/0226657 A1 to Bohe et al, describes a customer analytic record which includes the consumer’s digital consumption data along with the telecommunication data, the consumer’s behaviour profiles and customer’s attitude expressed through online surveys. All the available data, related to the consumer, is collected from a variety of sources for forming the data structure of the customer analytic record. Although, the present document relies on vast data sources for segmenting users, however, the present document does not specify any particular mobile device usage behaviour which can be identified for targeting a particular group of mobile device users.

Another United States Patent Application Publication Number 2014/0040017 A1 to Bafna et al, describes a method of mobile analytics for selling mobile applications in particular, over the mobile marketplace. The method includes monitoring the usage of mobile applications on the user’s mobile device and focuses on grouping users based on the usage of the mobile applications on the user’s mobile devices. Thereafter, the usage data related to the mobile applications is provided to the analytics group of marketing campaigns. The present document focus is limited to tracking only the mobile applications for determining usage behaviour of mobile users.

The above disclosed methods provide an aggregated profile of the customers and in some cases the mobile application usage of the customers to a group of mobile marketing companies. However, the existing methods do not provide any means of assessing the overall usage of mobile devices and thereafter determine such usage which is most likely representative of the user’s specific preference for any service or product. Therefore, there is a need amongst mobile marketing companies to be able to relate the overall mobile device usage to his actual preference for a product or service. This shall assist in better specifying the target users of their services or products. Thus, there is a need for a method and system of analyzing mobile device usage data from which we can also assess the user’s actual preference for services and products.

The following is a brief description of the embodiments as illustrated in the accompanying drawings. It is to be understood that the features illustrated in and described with reference to the drawings are not to be construed as limiting of the scope of the present disclosure. In the accompanying drawings:

Figure 1 illustrates an embodiment of the system for identifying a target set of mobile device users in accordance with the present disclosure.

Figure 2 illustrates another embodiment of the system for identifying a target set of mobile device users in accordance with the present disclosure.

A system for identifying a target set of mobile device users is disclosed. The disclosed system comprises of a receiving module configured to receive time series usage data for multiple usage parameters from multiple mobile devices of a group of users. The receiving module is further configured to receive user attitude data from the users of the group of users. The system for identifying a target set of mobile device users further comprises of a database configured to store the data received by the receiving module. Further, the system also comprises of an analytics module configured to analyze the usage data to identify clusters, such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on its priority rank. Further, the analytics module is configured to analyze each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster and to identify co-relations between the usage pattern and the user attitude data for a cluster. Further, the analytics module is configured to link the usage pattern to the attitude data on identifying a co-relation, and to store the usage pattern as a target set of mobile device users in the database.

It will be understood by those skilled in the art that the foregoing objects and the following description of the nature of invention are exemplary and explanatory of the invention and are not intended to be restrictive thereof.

For the purpose of promoting an understanding of the principles of the invention, reference will now be made to various alternative embodiments and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended, such alterations and further modifications in the illustrated method and system, and such further applications of the principles of the invention as illustrated therein being contemplated as would normally occur to one skilled in the art to which the invention relates.

It will be understood by those skilled in the art that the following description is exemplary and explanatory of the invention and are not intended to be restrictive thereof.

Many of the functional units described in this specification have been labeled as modules, in order to more particularly emphasize their implementation independence. For example, a module may be implemented as a hardware circuit comprising custom very large scale integration circuits or gate arrays, off-the-shelf semiconductors such as logic, chips, transistors, or the other discrete components. A module may also be implemented in programmable hardware devices such as field programmable gate arrays, programmable array logic, programmable logic devices or the like.

Modules may also be implemented in software for execution by various types of processors or microporocessors. An identified module of executable code may, for instance, comprise one or more physical or logical blocks of computer instructions which may, for instance, be organized as an object, procedure, or function. Nevertheless, the executable of an identified module need not be physically located together, but may comprise disparate instructions stored in different locations which, when joined together, comprise the module and achieve the stated purpose for the module.

Indeed, a module of executable code could be a single instruction, or many instructions, and may even be distributed over several different code segments, among different programs, and across several memory devices. Similarly, operational data may be identified and illustrated herein within modules, and may be embodied in any suitable form and organized within any suitable type of data structure. The operational data maybe collected as a single data set, or may be distributed over different locations including over different member disks, and may exist, at least partially, merely as electronic signals on a system or network.

Reference throughout this specification to “one embodiment” “an embodiment” or similar language means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the present invention. Thus, appearances of the phrase “in one embodiment”, “in an embodiment” and similar language throughout this specification may, but do not necessarily, all refer to the same embodiment.

A method and a system of identifying a target set of mobile device users are disclosed. The target set of mobile device users includes those mobile device users who are likely to be the consumers of a marketing company’s services or products. Such consumers once identified can be targeted by marketers by sending out the relevant marketing messages to them.

A method of identifying a target set of mobile device users comprises of collecting time series usage data for multiple usage parameters from mobile devices of a group of users, analyzing the usage data to identify clusters amongst the mobile device users and analyzing each identified cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster. Further, the method comprises of collecting user attitude data from each user of the group of users, identifying co-relations between the usage pattern and the user attitude data for a cluster and, on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users. The usage data and the attitude data is collected from each user from the group of users who participate in providing such details to a campaign of a marketing company. The attitude data can be collected from various sources such as personal surveys, social media surveys and various other known behavioral collection methods known in the art. Further, the attitude data may include all forms of psychographic data, such as personality, values, lifestyle etc. which marketers capture to understand their target audiences.

Such attitude data is collected and thereafter analyzed and stored as a set of data points for obtaining co-relations using the method taught in the present disclosure. In accordance with an embodiment, an example of data points representing the attitude data of a user is: “I am an extrovert”=5, “I like watching movies” =2 etc. The time series usage data which is collected from mobile devices of a group of users is the data which is generated by a user as he uses his mobile device for regular activities such as calling, texting, messaging, mobile device applications etc. or when visiting locations are tracked by the mobile device of the user. The mobile device may include but not limited to a smartphone, tablet, or any other known device capable of using internet or features such as mobile applications, web browsing etc. In accordance with an embodiment, the time series usage data is collected by an application installed on the mobile devices of the users. The collected usage data can be analyzed according to the teachings of the present disclosure on the mobile device and can be further uploaded on a server in communication with the mobile device. Such usage data is collected in time series at regular predefined uniform intervals of time such as after every 5 minutes, or at any other uniform interval of time which can be set up to an hour. Alternatively, a default time interval of 30 minutes can be set. The time series usage data is collected for multiple usage parameters of the usage of the mobile device such as calling patterns, texting patterns, location, mobile application usage, browser usage, battery status, data network status etc. In accordance with an embodiment, the time series usage data which is collected from the mobile devices may include but not limited to:
1. Calling pattern
(i) Number of minutes spent on a call
(ii). Nature of call-incoming or outgoing
(iii) Top receivers of call by minutes;
(iv). This is further converted to number of receivers who use 80% of all outbound minutes
(v). Top callers by minutes
This is further converted to number of callers who use 80% of all inbound minutes

2. Texting patterns
(i) Number of texts sent during the set time interval
(ii). Number of texts received during the set time interval
(iii). Top receivers of texts
This is further converted to number of recipients who receive 80% of all messages sent
(iv). Top senders of texts
This is further converted to number of senders who send 80% of all messages
(v). Special sender category of SMS
This category is determined using a supplied list of special SMS sender names, for example, SMS received from banks such as ‘LM-ICICIB’ or ‘DM-KOTAKB’, wherein the user has a financial relationship with a private bank, ‘ICICI’ or ‘KOTAK’. Further, examples may include SMS received from ‘AM-IRCTCi’, where the user is a ‘IRCTC’ website user.

3. Location
(i). Geographic Latitudinal and Longitudinal values of current location
This is further converted to a “type” of “point of interest” using a database containing simple table of point of interest, their “type” examples: ATM, mall, multiplex, school, gym) and exact location (latitudinal value, longitudinal value)
4. Mobile application usage
(i). All mobile applications that are running currently or were used recently on the mobile device of the users. This information may be aggregated over a day to determine
a.Total number of distinct mobile applications that were used
b. Top 5 mobile applications used
c. Number of mobile applications in the top 5 mobile application list that are present in a given list of top 100 mobile applications
d. Number of mobile applications in the top 5 mobile applications that are not present in a given list of top-100 applications
5. Browser usage
(i). Websites visiting during the set time interval. This information is aggregated over a day
(ii). Top 5 websites visited
(iii).Number of websites in the top 5-list that are present in a given list of Top-100 websites
(iv). Number of websites in the top -5 list that are not present in a given list of top-100 websites
6. Battery status
(i). Battery charge remaining (0-100%)
(ii). If charger is connected or not
7. Data network status
(i). Network signal strength (0-100%)
(ii). If Data Network is on or not
(iii). If Wifi Network is on or not

Further, the usage data is analyzed to identify clusters. The clusters of mobile devices created which generate similar usage data. Each of the usage parameters for which data point are collected from the mobile devices of a group of users, are assigned a preset priority rank. In accordance with an embodiment, the priority ranks are assigned by default to the usage parameters. An example of such default priority rank assignment, from high to low, is as follows:
(i). Location data points
(ii). Calling pattern data points
(iii). Mobile Application usage data points
(iv).Texting pattern data points
(v). Browser usage data points
(vi). Data Network status data points
(vii). Battery status data points

Based on the priority ranks assigned to each of the usage parameters, the collected usage data is analyzed to identify clusters amongst the mobile device users. Thereafter, the analyzed usage data is used to identify clusters by giving weightage to the usage parameters based on its priority rank. The priority ranks determine the weightage of each data point. In accordance with an embodiment, the usage data is continuously analyzed over a predefined period of time to determine clusters. Such predefined period of time may be set to be a minimum period of time such as two to four weeks.

Further, each cluster is analyzed to extract a usage pattern. The usage behavior is a common mobile device usage behavior exhibited by user’s mobile devices from the group of users. Such common usage behavior is identified for each of the clusters, and the description of the identified usage behavior is saved as usage patterns. In accordance with an embodiment, the description of the usage patterns may include but not limited to:

(i). A news reading mobile application is used more than twice a day
(ii). A social media mobile application is seen in top 5 mobile application list 90% of times
(iii). Cinema hall is visited more than once a week
(iv). Incoming call minutes to outbound call minutes ratio is 1.5
(v). 90% text messages are sent to contacts in top 5 messaged contact list
(vi). 50% battery charge is remaining 90% of times

In accordance with further embodiments of the present disclosure, the attitude data is collected from each user from the group of users, analyzed and stored or uploaded on the server on which the stored user pattern also resides. The attitude data analysis is done according to the known teachings of the art.

In accordance with further embodiments of the present disclosure, the method of identifying a target set of mobile device users comprises of identifying co-relations between the usage pattern and the user attitude data for a cluster. Co-relations between the usage pattern and the user attitude data are identified to detect if any similarities regarding the usage behaviour exists between the usage pattern of any cluster and the attitude data collected from the each mobile device user of the same cluster. If the usage behavior described by a usage pattern exhibits any similarity to the attitude data, a co-relation is said to be identified.

In accordance with further embodiments of the present disclosure, the method of identifying a target step of mobile device users further comprises of identifying new clusters, analyzing each new cluster to extract a usage pattern, and identifying co-relations between the usage pattern and the user attitude data for the new cluster. The new clusters are identified in order to make attempts to identify meaningful co-relations between the attitude data and the usage pattern of the clusters. The new clusters are identified amongst the mobile device users, by giving one or more usage parameters a new priority rank. Thereafter, the new cluster is identified by giving weightage to each usage parameter based on its new priority rank.

According to an embodiment, the new clusters are identified till a degree of similarity between the usage pattern and the user attitude data is achieved. As the case may be, no such similarity may exist, or in another case some degree of similarity may exist, or in another case a higher or an absolute similarity may exists between the usage pattern and the user attitude data. Such degree of similarity that should be achieved may be predefined, for example, a minimum 30% or 50% similarity in the mobile device usage behavior and the user attitude data must exist, for a co-relation to be identified.

In accordance with further embodiments of the present disclosure, the method of identifying a target set of mobile device users comprises of linking the usage pattern to the attitude data on identifying a co-relation. Further, the method comprises of saving the usage pattern as a target set of mobile device users. These saved usage pattern shall essentially represent the set of consumers, to be targeted by marketing companies, as the likely users of their product and services. In accordance with another embodiment of the invention, co-relations are identified for each cluster, and the usage pattern are saved for those clusters which exhibit a high degree of similarity with the user attitude data of those clusters. Multiple such usage patterns may be identified which exhibit a high degree of similarity, and thereafter are linked to the user attitude data and saved as multiple target sets of mobile device users. A high degree of similarity may be predefined, for example, an 70% degree of similarity may qualify a usage pattern to be saved as a target set of mobile device users.

A system for identifying a target set of mobile device users is disclosed. The disclosed system comprises of a receiving module configured to receive time series usage data for multiple usage parameters from multiple mobile devices of a group of users. The receiving module is further configured to receive user attitude data from the users of the group of users. The system further comprises of a database configured to store the data received by the receiving module. The system further comprises of an analytics module configured to analyze the usage data to identify clusters, such that each cluster is identified by giving weightage to each usage parameter based on its priority rank. Further, the analytics module is configured to analyze each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster. Further, the analytics module is configured to identify co-relations between the usage pattern and the user attitude data for a cluster, and on identifying a co-relation, to link the usage pattern to the attitude data and store the usage pattern as a target set of mobile device users in the database. Further, the database is configured to store the data received from the analytics module.

In accordance with an embodiment, the analytics module is further configured to identify new clusters by giving one or more usage parameters a new priority rank according to the teachings of the present disclosure. Further, the analytics module is configured to analyze each new cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that new cluster. Further, the analytics module is configured to identify co-relations between the usage pattern and the user attitude data for the new cluster and on identifying a co-relation, to link the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.

In accordance with an example, Figure 1 depicts a system (100) for identifying a target set of mobile device users in accordance with the teachings of the present disclosure. The system (100) comprises of a receiving module (200), a database (300) and an analytics module (400). The receiving module (200) and the analytics module (400) can be implemented either as a hardware circuit, a hardware programmable device or as software logic, or in the form of programmable microprocessors or by such other various means known in the art configured to perform the functions as taught in the present disclosure. The database (300) can be a memory device or storage implemented as hardware circuit or software logic, and is configured to store the data as received from the receiving module (200). Further, the analytics module (400) is in communication with the database (300) to access the data from therewith and perform the functions as disclosed in the teachings of the present disclosure, and also to store those usage patterns in the database (300) which are identified to be target set of mobile device users.

In accordance with an embodiment, the entire system (100) to identify a target set of mobile devices can reside on a central server accessed by the operators of the mobile marketing company. Further, the central server is in communication with the mobile devices of the users of the group of users and is configured to access the time series usage data from the mobile device of the users. Further, the attitude data collected from the marketing campaigns is received or uploaded on the receiving module (200) and stored on the database (300) on the central server for analysis. Alternatively, the said modules and the databases can reside on separate servers individually and are configured to remotely access data from each others through various communications means known in the art.

In accordance with another embodiment, Figure 2 shows an example system (100) for identifying a target set of mobile users in accordance with the teachings of the present disclosure. The system (100) comprises of a two receiving modules, receiving module 1 (200A) and receiving module 2 (200B), a database (300) and an analytics module (400). The receiving module 1 (200A) is configured to receive mobile usage data from a group of mobile device users and the receiving module 2 (200B) is configured to receive the attitude data from the group of users. The data thus collected separately by the two receiving modules, receiving module 1 (200A) and receiving module 2 (200B) are then uploaded on the database (300).

In accordance with another embodiment, the receiving module (200A) may reside on the mobile device of the user and is configured to access the time series usage data from the mobile devices. The receiving module (200A) in communication with the central server can then upload the collected usage data to the database (300) residing on the central server.

SPECIFIC EMBODIMENTS

A method of identifying a target set of mobile device users is disclosed. The method comprises of collecting time series usage data for multiple usage parameters from mobile devices of a group of users, analyzing the usage data to identify clusters; such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on its priority rank. The method further comprises of analyzing each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster. Further, the method comprises of collecting user attitude data from the users of the group of users, identifying co-relations between the usage pattern and the user attitude data for a cluster and, on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.

Such method(s) further comprising of identifying new clusters by giving one or more usage parameters a new priority rank and the new cluster is identified by giving weightage to each usage parameter based on its new priority rank. Further such method comprising of analyzing each new cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that new cluster. Further, such method comprising of identifying co-relations between the usage pattern and the user attitude data for the new cluster; and on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.

A system for identifying a target set of mobile device users is disclosed. The disclosed system comprises of a receiving module configured to receive time series usage data for multiple usage parameters from multiple mobile devices of a group of users. The receiving module is further configured to receive user attitude data from the users of the group of users. The system further comprises of a database configured to store the data received by the receiving module. Further, the system also comprises of an analytics module configured to analyze the usage data to identify clusters, such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on its priority rank. Further, the analytics module is configured to analyze each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster. Further, the analytics module is configured to identify co-relations between the usage pattern and the user attitude data for a cluster, and on identifying a co-relation, link the usage pattern to the attitude data and store the usage pattern as a target set of mobile device users in the database.

Such system(s) wherein the analytics module is further configured to identify new clusters by giving one or more usage parameters a new priority rank and the new cluster is identified by giving weightage to each usage parameter based on its new priority rank. Further, the analytics module is configured to analyze each new cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that new cluster. Further, the analytics module is configured to identify co-relations between the usage pattern and the user attitude data for the new cluster and on identifying a co-relation, link the usage pattern to the attitude data and store the usage pattern as a target set of mobile device users in the database.

The disclosed method and system introduces a practical way to use smartphones to describe the consumers. Usage patterns as taught in the present disclosure represent meaningful co-relations between smartphone usage behaviour and the set of consumers, which marketing companies emphasize on based on the attitude and the psychographic data collected from the users. Such users represent the target mobile users for mobile marketing companies. Thus, usage patterns can be used to address a set of people for business actions such as sending marketing messages by appropriate means. Such appropriate means would represent the most common characteristic of the target set of mobile device users. By creating the usage patterns as described in the present disclosure, the mobile marketing companies can identify their consumers by simply installing the identified usage pattern on the mobile devices of their users and detect any likely match on the mobile device itself. Therefore, no mobile data of the consumers is externally shared and privacy of an individual user’s mobile device data is always ensured.

Claims

A method of identifying a target set of mobile device users comprising:
collecting time series usage data for multiple usage parameters from mobile devices of a group of users;
analyzing the usage data to identify clusters; such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on it’s priority rank;
analyzing each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster;
collecting user attitude data from the users of the group of users;
identifying co-relations between the usage pattern and the user attitude data for a cluster; and
on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.
A method as claimed in claim 1 further comprising:
identifying new clusters by giving one or more usage parameters a new priority rank and the new cluster is identified by giving weightage to each usage parameter based on its new priority rank;
analyzing each new cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that new cluster;
identifying co-relations between the usage pattern and the user attitude data for the new cluster; and
on identifying a co-relation, linking the usage pattern to the attitude data and saving the usage pattern as a target set of mobile device users.
A system for identifying a target set of mobile device users comprising:
a receiving module configured to receive time series usage data for multiple usage parameters from multiple mobile devices of a group of users; the receiving module further configured to receive user attitude data from the users of the group of users;
a database configured to store the data received by the receiving module;
an analytics module configured to:
- analyze the usage data to identify clusters, such that each usage parameter is given a priority rank and a cluster is identified by giving weightage to each usage parameter based on its priority rank;
- analyze each cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that cluster;
- identify co-relations between the usage pattern and the user attitude data for a cluster;
- on identifying a co-relation, link the usage pattern to the attitude data; and
- store the usage pattern as a target set of mobile device users in the database.
A system for identifying a target set of mobile device users as claimed in claim 4 wherein the analytics module further configured to:
- identify new clusters by giving one or more usage parameters a new priority rank and the new cluster is identified by giving weightage to each usage parameter based on its new priority rank;
- analyze each new cluster to extract a usage pattern, the usage pattern describing usage behavior of a user of a mobile device from that new cluster;
- identify co-relations between the usage pattern and the user attitude data for the new cluster; and
- on identifying a co-relation, link the usage pattern to the attitude data and store the usage pattern as a target set of mobile device users in the database.