Method and apparatus to identify influencers

• US 9,946,975 B2
• Filed: 08/20/2008
• Issued: 04/17/2018
• Est. Priority Date: 08/24/2007
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Paul, Deirdre, Volinsky, Christopher
• Primary Examiner(s): Gart, Matthew S
• Assistant Examiner(s): YOUNG, ASHLEY YA-SHEH
• Application Number:
  US12/195,246
• Publication Number:
  US 20090063254A1
• Time in Patent Office: 3,527 Days
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

A method and apparatus for identifying influencers are disclosed. For example, the method obtains a list of customers, and determines a social network for each of the customers. The method selects one or more attributes to be used for predicting a measure of influence for each of the customers, and determines one or more influencers in the social network by using the one or more attributes.

Claims

1. A method for identifying an influencer, comprising:

determining, via a processor, a social network for each of a plurality of customers based upon a group of users who have exchanged communications with each of the plurality of customers, wherein the determining the social network for each of the plurality of customers comprises;

defining a neighborhood for each of the plurality of customers; and

determining a strength of relationships of each of the plurality of customers with members of the neighborhood of the customer;

truncating the neighborhood for each of the plurality of customers by selecting k neighbors of each of the plurality of customers, wherein the k is a predetermined integer;

wherein the truncating the neighborhood is performed by selecting a most important k neighbors using the strength of relationships;

recursively finding a respective neighborhood for each of the remaining members in the neighborhood for a pre-determined number of times; and

defining the social network for each of the plurality of customers as the neighborhood that comprises the truncated neighborhood of the customer and the respective neighborhood that is recursively found;

selecting, via the processor, an attribute associated with communications among the plurality of customers to be used for predicting a measure of influence for each of the plurality of customers, wherein the attribute comprises a willingness of a customer to adopt new services;

building, via the processor, a model based upon the attribute associated with the communications among the plurality of customers that is selected, the social network of each of the plurality of customers and a sequence of product adoption events of each of the plurality of customers for predicting the measure of influence for each of the plurality of customers, wherein the model comprises a statistical model, wherein the statistical model comprises a linear regression model, wherein the sequence of product adoption events includes each of the plurality of customers first adopting a new service and subsequently users in the group of users who have exchanged communications with each of the plurality of customers also adopting the new service; and

determining, via the processor, the influencer in the social network by applying the model.

2. The method of claim 1, further comprising:

providing an action to be taken for the influencer.

3. The method of claim 2, wherein the action comprises a directing marketing action.

4. A non-transitory computer-readable medium storing instructions which, when executed by a processor, cause the processor to perform operations for identifying an influencer, the operations comprising:

determining a social network for each of a plurality of customers based upon a group of users who have exchanged communications with each of the plurality of customers, wherein the determining the social network for each of the plurality of customers comprises;

defining a neighborhood for each of the plurality of customers; and

determining a strength of relationships of each of the plurality of customers with members of the neighborhood of the customer;

truncating the neighborhood for each of the plurality of customers by selecting k neighbors of each of the plurality of customers, wherein the k is a predetermined integer;

wherein the truncating the neighborhood is performed by selecting a most important k neighbors using the strength of relationships;

recursively finding a respective neighborhood for each of the remaining members in the neighborhood for a pre-determined number of times; and

defining the social network for each of the plurality of customers as the neighborhood that comprises the truncated neighborhood of the customer and the respective neighborhood that is recursively found;

selecting an attribute associated with communications among the plurality of customers to be used for predicting a measure of influence for each of the plurality of customers, wherein the attribute comprises a willingness of a customer to adopt new services;

building a model based upon the attribute associated with the communications among the plurality of customers that is selected, the social network of each of the plurality of customers and a sequence of product adoption events of each of the plurality of customers for predicting the measure of influence for each of the plurality of customers, wherein the model comprises a statistical model, wherein the statistical model comprises a linear regression model, wherein the sequence of product adoption events includes each of the plurality of customers first adopting a new service and subsequently users in the group of users who have exchanged communications with each of the plurality of customers also adopting the new service; and

determining the influencer in the social network by applying the model.

5. The non-transitory computer-readable medium of claim 4, wherein the operations further comprise:

providing an action to be taken for the influencer.

6. The non-transitory computer-readable medium of claim 5, wherein the action comprises a directing marketing action.

7. An apparatus for identifying an influencer, comprising:

a processor; and

a computer-readable medium storing instructions which, when executed by the processor, cause the processor to perform operations, the operations comprising;

determining a social network for each of a plurality of customers based upon a group of users who have exchanged communications with each of the plurality of customers, wherein the determining the social network for each of the plurality of customers comprises;

defining a neighborhood for each of the plurality of customers; and

determining a strength of relationships of each of the plurality of customers with members of the neighborhood of the customer;

truncating the neighborhood for each of the plurality of customers by selecting k neighbors of each of the plurality of customers, wherein the k is a predetermined integer;

wherein the truncating the neighborhood is performed by selecting a most important k neighbors using the strength of relationships;

recursively finding a respective neighborhood for each of the remaining members in the neighborhood for a pre-determined number of times; and

defining the social network for each of the plurality of customers as the neighborhood that comprises the truncated neighborhood of the customer and the respective neighborhood that is recursively found;

selecting an attribute associated with communications among the plurality of customers to be used for predicting a measure of influence for each of the plurality of customers, wherein the attribute comprises a willingness of a customer to adopt new services;

building a model based upon the attribute associated with the communications among the plurality of customers that is selected, the social network of each of the plurality of customers and a sequence of product adoption events of each of the plurality of customers for predicting the measure of influence for each of the plurality of customers, wherein the model comprises a statistical model, wherein the statistical model comprises a linear regression model, wherein the sequence of product adoption events includes each of the plurality of customers first adopting a new service and subsequently users in the group of users who have exchanged communications with each of the plurality of customers also adopting the new service; and

determining the influencer in the social network by applying the model.

8. The apparatus of claim 7, wherein the operations further comprise:

providing an action to be taken for the influencer.