Goal based conversational serendipity inclusion

• US 10,114,890 B2
• Filed: 06/30/2015
• Issued: 10/30/2018
• Est. Priority Date: 06/30/2015
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Baughman, Aaron K., Kellogg, Wendy A., Kozloski, James R., Salapura, Valentina
• Primary Examiner(s): Casanova, Jorge A
• Application Number:
  US14/788,156
• Publication Number:
  US 20170004207A1
• Time in Patent Office: 1,218 Days
• CPC Class Codes: G06F 16/35 Clustering; Classification

Abstract

A method for injecting information into a deliberation includes extracting at least one concept from a plurality of statements asserted by at least one participant of a group, correlating the at least one concept to a goal of the deliberation, tagging at least one of the plurality of statements with an identifier of a participant who asserted the statement, clustering a plurality of participants of the group according to a slant on each of the at least one concept, searching, using the at least one concept, tagged statements and a cluster of participants, for new information or concepts that negate at least one slant that is not in agreement with the goal, and injecting the new information into the deliberation.

Claims

1. A method of operating an electronic meeting assistant injecting information into a deliberation over time comprising:

extracting, by an input module of the electronic meeting assistant system, at least one concept from a plurality of statements asserted by at least one participant of a group;

correlating, by a correlation module of the electronic meeting assistant system, the at least one concept to a goal of the deliberation, wherein the goal is consensus of the group;

tagging, by a tagging module of the electronic meeting assistant, at least one of the plurality of statements with an identifier of a participant who asserted the statement;

clustering, by a clustering module of the electronic meeting assistant, a plurality of participants of the group according to a slant on each of the at least one concept;

searching, by a retrieval system of the electronic meeting assistant, using the at least one concept, the statements tagged by the tagging module and a set of the participants clustered according to a certain slant that is not in agreement with the goal, for new information or concepts that negate the certain slant; and

injecting, by an injector module of the electronic meeting assistant, the new information into the deliberation at a rate determined by a degree of negation of the new information or concepts towards the goal, wherein the degree of negation is updated after each injection of the new information and wherein the rate of injection is selected to maintain a rate of convergence to the goal.

2. The method of claim 1, wherein the injection is individualized to each group participant.

3. The method of claim 1, wherein searching further comprises amplifying information or concepts that support the goal.

4. The method of claim 1, where searching further comprises ranking a salience of the at least one concept based on the support of the goal.

5. The method of claim 1, where searching further comprises ranking a salience of the new information based on the support of the goal.

6. The method of claim 1, further comprising scoring a serendipity of the new information based on correlation to the goal.

7. The method of claim 6, further comprising searching for a serendipity concept in a topic space based on a measure of orthogonality from a plurality of topics discussed in the deliberation.

8. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method of operating an electronic meeting assistant injecting information into a deliberation over time comprising:

extracting at least one concept from a plurality of statements asserted by at least one participant of a group;

correlating the at least one concept to a goal of the deliberation, wherein the goal is consensus of the group;

tagging at least one of the plurality of statements with an identifier of a participant who asserted the statement;

clustering a plurality of participants of the group according to a slant on each of the at least one concept;

searching, using the at least one concept, tagged statements and a cluster of participants, for new information or concepts that negate at least one slant that is not in agreement with the goal; and

injecting the new information into the deliberation at a rate determined by a degree of negation of the new information or concepts toward the goal, wherein the degree of negation is updated after each injection of the new information and wherein the rate of injection is selected to maintain a rate of convergence to the goal.

9. The computer program product of claim 8, wherein the injection is individualized to each group participant.

10. The computer program product of claim 8, wherein the injection is presented using a display.

11. The computer program product of claim 10, further comprising reducing an oscillation in the deliberation by updating the degree of negation.

12. The computer program product of claim 8, further comprising amplifying information or concepts that support the goal during a search.

13. The computer program product of claim 8, further comprising ranking, during a search, a salience of the at least one concept based on the support of the goal.

14. The computer program product of claim 8, further comprising ranking, during a search, a salience of the new information based on the support of the goal.

15. The computer program product of claim 8, further comprising scoring a serendipity of the new information based on correlation to the goal.

16. The computer program product of claim 15, further comprising searching for a serendipity concept in a topic space based on a measure of orthogonality from a plurality of topics discussed in the deliberation.