Assessment of communication strengths of individuals from electronic messages

• US 20030055711A1
• Filed: 07/02/2001
• Published: 03/20/2003
• Est. Priority Date: 07/02/2001
• Status: Active Grant

• Current Assignee: The Procter & Gamble Company
• Original Assignee: The Procter & Gamble Company
• Inventors: Doherty, Michael Emmett
• Granted Patent: US 7,143,054 B2
• Time in Patent Office: Days
• US Class Current: 705/10
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0203 Market surveys; Market polls
• CPC Class Codes: G06Q 30/0204 Market segmentation

Abstract

An apparatus, program product, and method analyze electronic messages directed between individuals to quantitatively assess the relative communication capabilities of such individuals. A quantitative value referred to as “communication strength” is determined for one or more individuals from a plurality of individuals based upon analysis of the electronic messaging activities of such individual(s), and is subsequently used to select an individual from among the plurality of individuals, e.g., for the purpose of selecting a target for a viral marketing encounter, a market research project, etc. The communication strength of an individual is determined by modeling a plurality of conversations identified through analysis of a plurality of electronic messages generated by the plurality of individuals, where each conversation is associated with a talker and a listener from the plurality of individuals, and is useful in determining a level of messaging activity between the talker and listener associated with such conversation.

Claims

1. A method of assessing communication strength of individuals that participate in electronic communications, the method comprising:

(a) analyzing a plurality of electronic messages generated by a plurality of individuals to identify a plurality of conversations, wherein each conversation is associated with a talker and a listener, and wherein each talker and listener identified by a conversation is selected from the plurality of individuals;

(b) determining communication strengths of at least a subset of the plurality of individuals by modeling the plurality of conversations; and

(c) selecting an individual from the plurality of individuals based upon the communication strength thereof.

2. The method of claim 1, further comprising harvesting the plurality of electronic messages based upon a domain of interest.

3. The method of claim 2, wherein harvesting the plurality of electronic messages includes searching at least one message archive to identify the plurality of electronic messages and creating a local message archive from which information regarding the plurality of electronic messages can be obtained during analysis.

4. The method of claim 3, wherein the message archive from which the electronic messages are harvested includes at least one of a news archive, a forum archive, a mailing list archive, a chat archive, an instant messaging archive, a telephone record archive and an email archive, and wherein each of the plurality of messages is selected from the group consisting of a message post, an email, a chat post, an instant message, and a telephone record.

5. The method of claim 2, further comprising selecting a study domain from the domain of interest based upon at least one study parameter.

6. The method of claim 1, further comprising determining, from the plurality of conversations, a plurality of activity indicators, each activity indicator associated with first and second individuals from the plurality of individuals, and each activity indicator representing a level of messaging activity directed from the first individual associated with such activity indicator to the second individual associated with such activity indicator.

7. The method of claim 6, wherein each activity indicator comprises a conversation count representative of the number of conversations directed from the first individual associated with such activity indicator to the second individual associated with such activity indicator.

8. The method of claim 6, further comprising:

(a) identifying a first subset of the plurality of individuals as talkers for the purposes of the analysis based upon a talker criterion; and

(b) identifying a second subset of the plurality of individuals as listeners for the purposes of the analysis based upon a listener criterion.

9. The method of claim 8, wherein modeling the plurality of conversations includes populating a matrix with activity indicators associated with messaging activity directed from the individuals identified as talkers based upon the talker criterion.

10. The method of claim 9, wherein populating the matrix includes arranging in different areas of the matrix activity indicators associated with messaging activity directed to individuals identified as talkers based upon the talker criterion, and activity indicators associated with messaging activity directed to individuals identified as listeners based upon the listener criterion.

11. The method of claim 9, wherein determining the communication strengths includes simultaneously solving talking strengths and listening strengths for at least a subset of the plurality of individuals using the matrix.

12. The method of claim 11, wherein simultaneously solving talking strengths and listening strengths includes solving an eigenvalue problem using the relationships:

v=Wuu=W^Tv

wherein

W=the matrix;

W^T=the transpose of the matrix;

u=a vector of talking strengths; and

v=a vector of listening strengths.

13. The method of claim 6, further comprising scaling each activity indicator using a wearout factor prior to determining the communication strengths.

14. The method of claim 13, wherein determining the communication strengths determines a first set of communication strengths, and wherein the method further comprises determining a second set of communication strengths after scaling the activity indicators using a second wearout factor.

15. The method of claim 14, further comprising generating a diagram that contrasts the first and second sets of communication strengths.

16. The method of claim 15, wherein the diagram includes first and second orthogonal axes respectively representing the first and second sets of communication strengths, and wherein the diagram includes a plurality of data points, each representing the communication strengths in each of the first and second sets of communication strengths determined for an individual from the plurality of individuals.

17. The method of claim 16, wherein the diagram further includes bubbles representing groups of individuals having like communication strength characteristics, wherein the size of each bubble is representative of the number of individuals within the group of individuals represented by such bubble.

18. The method of claim 1, wherein modeling the plurality of conversations includes expressing from a model the propositions that a strong talker talks frequently to strong listeners, and that a strong listener listens frequently to strong talkers.

19. The method of claim 1, wherein determining the communication strength for an individual includes determining at least one of a talking strength and a listening strength for such individual.

20. The method of claim 19, wherein selecting an individual from the plurality of individuals based upon the communication strength thereof includes identifying the individual as a connector based at least upon the talking strength of such individual.

21. The method of claim 20, further comprising contacting the individual to perform a marketing activity therewith.

22. The method of claim 21, further comprising initiating an encounter with the individual in a viral marketing context.

23. The method of claim 21, further comprising performing market research with the individual.

24. A method of conducting a viral marketing activity, the method comprising:

(a) identifying a connector from a plurality of individuals by assessing communication strengths for the plurality of individuals, including determining the communication strengths through analysis of a plurality of electronic messages generated by the plurality of individuals and modeling of a plurality of conversations identified through analysis of the plurality of electronic messages; and

(b) initiating an encounter with the connector.

25. An apparatus, comprising:

(a) a memory; and

(b) a program resident in the memory and configured to analyze a plurality of electronic messages generated by a plurality of individuals to identify a plurality of conversations, determine communication strengths of at least a subset of the plurality of individuals by modeling the plurality of conversations, and select an individual from the plurality of individuals based upon the communication strength thereof, wherein each conversation is associated with a talker and a listener, and wherein each talker and listener identified by a conversation is selected from the plurality of individuals.

26. The apparatus of claim 25, further comprising a local message archive resident in the memory, wherein the program is further configured to harvest the plurality of electronic messages based upon a domain of interest by searching at least one other message archive to identify the plurality of electronic messages and storing in the local message archive information regarding the plurality of electronic messages, wherein the other message archive includes at least one of a news archive, a forum archive, a mailing list archive, a chat archive, an instant messaging archive, a telephone record archive and an email archive, and wherein each of the plurality of messages is selected from the group consisting of a message post, an email, a chat post, an instant message, and a telephone record.

27. The apparatus of claim 25, wherein the program is further configured to determine, from the plurality of conversations, a plurality of activity indicators, each activity indicator associated with first and second individuals from the plurality of individuals, and each activity indicator representing a level of messaging activity directed from the first individual associated with such activity indicator to the second individual associated with such activity indicator.

28. The apparatus of claim 27, wherein each activity indicator comprises a conversation count representative of the number of conversations directed from the first individual associated with such activity indicator to the second individual associated with such activity indicator.

29. The apparatus of claim 27, wherein the program is further configured to identify a first subset of the plurality of individuals as talkers for the purposes of the analysis based upon a talker criterion, and to identify a second subset of the plurality of individuals as listeners for the purposes of the analysis based upon a listener criterion.

30. The apparatus of claim 29, wherein the program is configured to model the plurality of conversations by populating a matrix with activity indicators associated with messaging activity directed from the individuals identified as talkers based upon the talker criterion, and to determine the communication strengths by simultaneously solving talking strengths and listening strengths for at least a subset of the plurality of individuals using the matrix.

31. The apparatus of claim 27, wherein the program is further configured to scale each activity indicator using a wearout factor prior to determining the communication strengths.

32. The apparatus of claim 31, wherein the program is configured to determine the communication strengths by determining first and second sets of communication strengths after scaling the activity indicators by first and second wearout factors, respectively, and wherein the program is further configured to generate a diagram that contrasts the first and second sets of communication strengths.

33. The apparatus of claim 25, wherein the program is configured to model the plurality of conversations by expressing from a model the propositions that a strong talker talks frequently to strong listeners, and that a strong listener listens frequently to strong talkers.

34. The apparatus of claim 25, wherein the program is configured to determine the communication strength for an individual by determining at least one of a talking strength and a listening strength for such individual, and wherein the program is configured to select an individual from the plurality of individuals based upon the communication strength thereof by identifying the individual as a connector based at least upon the talking strength of such individual.

35. A program product, comprising:

(a) a program configured to analyze a plurality of electronic messages generated by a plurality of individuals to identify a plurality of conversations, determine communication strengths of at least a subset of the plurality of individuals by modeling the plurality of conversations, and select an individual from the plurality of individuals based upon the communication strength thereof, wherein each conversation is associated with a talker and a listener, and wherein each talker and listener identified by a conversation is selected from the plurality of individuals; and

(b) a signal bearing medium bearing the program.

36. The program product of claim 35, wherein the signal bearing medium includes at least one of a transmission medium and a recordable medium.