Data clustering and user modeling for next-best-action decisions

• US 10,453,083 B2
• Filed: 10/22/2015
• Issued: 10/22/2019
• Est. Priority Date: 05/16/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Herman, Norbert, Lambert, Daniel T.
• Primary Examiner(s): Spieler, William
• Application Number:
  US14/919,852
• Publication Number:
  US 20160042372A1
• Time in Patent Office: 1,461 Days
• Field of Search: 707737, 707825, 707706, 707736, 707790, 707722, 707769, 707749, 705 1453, 706 52, 382118
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/955 using information identifiers, e.g. uniform resource locators [URL]
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06Q 10/067 Enterprise or organisation modelling
• CPC Class Codes: G06Q 30/0204 Market segmentation
• CPC Class Codes: G06Q 30/0224 based on user history
• CPC Class Codes: G06Q 30/0255 based on user history
• CPC Class Codes: G06Q 50/01 Social networking

Abstract

Embodiments herein provide data clustering and user modeling for next-best-action decisions. Specifically, a modeling tool is configured to: receive indicators within unstructured social data from a plurality of users; analyze the unstructured social data of each of the plurality of users to assign a set of feature vectors to each of the plurality of users, each feature vector corresponding to one or more personality characteristics of each of the plurality of users; and analyze the feature vectors to identify two or more users from the plurality of users sharing a set of similar feature vectors. The modeling tool is further configured to: group the two or more users from the plurality of users sharing the set of similar feature vectors to form a cluster; identify attributes of the cluster; and input the attributes of the cluster into a predictive model to determine an offer corresponding to the cluster.

Claims

1. A method for targeting communications to a user, the method comprising the computer-implemented steps of:

receiving, by at least one computer device, unstructured social data of a plurality of users, the unstructured social data comprising one or more indicators of sentiment, personality, and current emotional state of specific users based on a set of words used by the specific users within the unstructured social data;

analyzing, by the at least one computer device, the unstructured social data created by each user of the plurality of users to reveal a personality of the user by, for each user of the plurality of users, automatically assigning a scoring value to each of a plurality of feature vectors, which are associated with the user, based on the set of words of the one or more indicators used by the user within the unstructured social data generated by the user, each of the set of feature vectors corresponding to one or more personality characteristics of the user that include at least one of;

a learning style, a socioeconomic class, or a personality trait of each of the plurality of users;

identifying, by the at least one computer device, attributes of each cluster of a plurality of clusters formed from two or more users from the plurality of users having common personality characteristics based on the feature vectors shared by users grouped in the cluster;

inputting, by the at least one computer device, the attributes of the cluster identified from the common personality characteristics into a predictive model to automatically determine a commercial offer that is tailored to the cluster based on the attributes of the users in the cluster and to automatically determine a second commercial offer that is tailored to a different cluster based on the attributes of the users in the cluster; and

forwarding, by the at least one computer device, the commercial offer that is tailored to the cluster to every user in the cluster and the second commercial offer that is tailored to the different cluster to every user in the different cluster.

2. The method according to claim 1, further comprising combining multiple feature vectors from the set of feature vectors to generate a set of higher-level feature vectors.

3. The method according to claim 1, further comprising receiving structured data of the plurality of users.

4. The method according to claim 3, the structured data comprises a history of past transactions.

5. The method according to claim 1, the analyzing the set of feature vectors comprising:

assigning a value to each of the set of feature vectors for each of the plurality of users, the value identifying a frequency and recentness of one or more indicators that indicate one or more personality characteristics corresponding to each of the set of feature vectors;

grouping each of the set of feature vectors based on the assigned value; and

plotting a graphical representation of the set of feature vectors for each of the plurality of users based on the grouping.

6. The method according to claim 5, further comprising determining a difference between graphical representations of the two or more users from the plurality of users.

7. The method according to claim 1, further comprising selecting the commercial offer for the cluster from a predetermined selection of commercial offers.

8. The method according to claim 7, further comprising:

receiving a response corresponding to the commercial offer from a responding user; and

changing the grouping of the responding user from the cluster to another cluster based on the response to the commercial offer.

9. A computer system for targeting communications to a user, the system comprising:

at least one processing unit;

memory operably associated with the at least one processing unit; and

a modeling tool storable in memory and executable by the at least one processing unit, the modeling tool comprising;

an analyzing component configured to;

receive unstructured social data of a plurality of users, the unstructured social data comprising one or more indicators of sentiment, personality, and current emotional state of specific users based on a set of words used by the specific users within the unstructured social data; and

analyze the unstructured social data created by each user of the plurality of users to reveal a personality of the user by, for each user of the plurality of users, automatically assigning a scoring value to each of a plurality of feature vectors, which are associated with the user, based on the set of words of the one or more indicators used by the user within the unstructured social data generated by the user, each of the set of feature vectors corresponding to one or more personality characteristics of the user that include at least one of;

a learning style, a socioeconomic class, or a personality trait of each of the plurality of users;

a clustering component configured to identify attributes of each cluster of a plurality of clusters formed from two or more users from the plurality of users having common personality characteristics based on the feature vectors shared by users grouped in the cluster; and

an offering component configured to input the attributes of the cluster identified from the common personality characteristics into a predictive model to automatically determine a commercial offer that is tailored to the cluster based on the attributes of the users in the cluster and a second commercial offer that is tailored to a different cluster based on the attributes of the users in the cluster and to forward the commercial offer that is tailored to the cluster to every user in the cluster and the second commercial offer that is tailored to the different cluster to every user in the different cluster.

10. The computer system according to claim 9, the analyzing component further configured to combine multiple feature vectors from the set of feature vectors to generate a set of higher-level feature vectors.

11. The computer system according to claim 9, the analyzing component further configured to receive structured data of the plurality of users, wherein the structured data comprises a history of past transactions.

12. The computer system according to claim 9, the analyzing component further configured to:

assign a value to each of the set of feature vectors for each of the plurality of users, the value identifying a frequency and recentness of one of more indicators that indicate one or more personality characteristics corresponding to each of the set of feature vectors;

group each of the set of feature vectors based on the assigned value;

plot a graphical representation of the set of feature vectors for each of the plurality of users based on the grouping; and

determine a difference between graphical representations of the two or more users from the plurality of users.

13. The computer system according to claim 9, the offering component further configured to select the commercial offer for the cluster from a predetermined selection of commercial offers.

14. The computer system according to claim 13, the offering component further configured to:

receive a response corresponding to the commercial offer from a responding user; and

change the grouping of the responding user from the cluster to another cluster based on the response to the commercial offer.

15. A computer-readable storage medium storing computer instructions, which when executed, enables a computer system for targeting communications to a user, the computer instructions comprising:

receiving unstructured social data of a plurality of users, the unstructured social data comprising one or more indicators of sentiment, personality, and current emotional state of specific users based on a set of words used by the specific users within the unstructured social data;

analyzing the unstructured social data created by each user of the plurality of users to reveal a personality of the user by, for each user of the plurality of users, automatically assigning a scoring value to each of a plurality of feature vectors, which are associated with the user, based on the set of words of the one or more indicators used by the user within the unstructured social data generated by the user, each of the set of feature vectors corresponding to one or more personality characteristics of the user that include at least one of;

a learning style, a socioeconomic class, or a personality trait of each of the plurality of users;

identifying attributes of each cluster of a plurality of clusters formed from two or more users from the plurality of users having common personality characteristics based on the feature vectors shared by users grouped in the cluster;

inputting the attributes of the cluster identified from the common personality characteristics into a predictive model to automatically determine a commercial offer that is tailored to the cluster based on the attributes of the users in the cluster and to automatically determine a second commercial offer that is tailored to a different cluster based on the attributes of the users in the cluster; and

forwarding, by the at least one computer device, the commercial offer that corresponds to the cluster to every user in the cluster and the second commercial offer that is tailored to the different cluster to every user in the different cluster.

16. The computer-readable storage medium according to claim 15, the computer instructions further comprising combining multiple feature vectors from the set of feature vectors to generate a set of higher-level feature vectors.

17. The computer-readable storage medium according to claim 15, further comprising computer instructions for receiving structured data of the plurality of users, wherein the structured data comprises a history of past transactions.

18. The computer-readable storage medium according to claim 15, the computer instructions for analyzing further comprising:

assigning a value to each of the set of feature vectors for each of the plurality of users, the value identifying a frequency and recentness of one of more indicators that indicate one or more personality characteristics corresponding to each of the set of feature vectors;

grouping each of the set of feature vectors based on the assigned value;

plotting a graphical representation of the set of feature vectors for each of the plurality of users based on the grouping; and

determining a difference between graphical representations of the two or more users from the plurality of users.

19. The computer-readable storage medium according to claim 15, further comprising computer instructions for selecting the commercial offer for the cluster from a predetermined selection of commercial offers.

20. The computer-readable storage medium according to claim 19, further comprising computer instructions for:

receiving a response corresponding to the commercial offer; and

changing the grouping of the responding user from the cluster to another cluster based on the response to the commercial offer.