System and method for modifying new message retransmission within a system for harvesting community knowledge

• US 7,246,121 B2
• Filed: 10/02/2002
• Issued: 07/17/2007
• Est. Priority Date: 10/02/2002
• Status: Expired due to Fees

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Lukose, Rajan Mathew, Adar, Eytan, Sengupta, Caesar, Tyler, Joshua Rogers
• Primary Examiner(s): Wong; Don
• Assistant Examiner(s): SHECHTMAN, CHERYL MARIA
• Application Number:
  US10/262,717
• Publication Number:
  US 20040068499A1
• Time in Patent Office: 1,749 Days
• Field of Search: 707/2, 707/9, 707/3, 707/7, 709/203, 709206-232
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/107 Computer-aided management of electronic mailing [e-mailing]
• CPC Class Codes: H04L 51/214 using selective forwarding
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34
• CPC Class Codes: H04L 69/329 in the application layer [OSI layer 7]
• CPC Class Codes: H04L 9/40 Network security protocols
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99937 Sorting
• CPC Class Codes: Y10S 707/99939 Privileged access

Abstract

A system and method is disclosed for knowledge management. The method discloses the elements of: generating a client profile on a client computer connected to a network; storing the profile on the computer; receiving a new message; scoring the new message with respect to the client profile; counting a number of message responses to the new message; and re-transmitting the new message to a first sub-set of client computers on the network, based on the response count. The system discloses means for implementing the method.

Claims

1. A method for knowledge management, comprising:

generating a client profile on a client computer connected to a network;

storing the profile on the computer;

transmitting a new message;

scoring, at the client computer, the new message with respect to the client profile;

counting a number of message responses from client computers to the new message; and

re-transmitting the new message to a first sub-set of client computers on the network, based on the response count.

2. The method of claim 1 wherein:

the generating element includes generating the client profile on the client computer connected to a peer-to-peer network; and

the re-transmitting element includes re-transmitting the new message on the peer-to-peer network.

3. The method of claim 1 wherein:

new messages are any messages received by the client computer; and

message responses are messages transmitted over the network responding to one or more new messages.

4. The method of claim 1, wherein the retransmitting element includes:

delaying re-transmission of the new message to the first sub-set of client computers for a randomized period of time, if the score exceeds a predetermined threshold.

5. The method of claim 1, wherein the retransmitting element includes:

generating proxy-profiles for a set of client computers on the network;

scoring the new message with respect to the proxy profiles; and

re-transmitting the new message to those client computers in the set of client computers which have proxy scores higher than a predetermined threshold.

6. The method of claim 1 wherein the counting element includes:

associating an actual response count attribute with the new message;

observing actual message responses over the network; and

incrementing the actual response count attribute for the new message each time one of the actual message responses corresponds to the new message.

7. The method of claim 6, further including:

waiting for a predetermined time period;

retrieving the actual response count after the predetermined time period; and

re-transmitting the new message to a second sub-set of client computers on the network, if the actual response count remains less than a predetermined minimum response count.

8. The method of claim 7:

wherein the second sub-set does not overlap the first sub-set.

9. The method of claim 7 further comprising:

repeating the waiting and re-transmitting elements with differing sub-sets of client computers on the network, until the actual response count reaches the predetermined minimum response count.

10. The method of claim 7 further comprising:

repeating the waiting and re-transmitting elements with differing sub-sets of client computers, until the new message has been re-transmitted to all of the client computers on the network.

11. The method of claim 6 wherein the re-transmitting element includes:

reducing a number of client computers in the first sub-set, if a predetermined minimum response count minus the actual response count is less than a first predetermined number; and

increasing the number of client computers in the first sub-set, if the predetermined minimum response count minus the actual response count is greater than a second predetermined number.

12. The method of claim 1:

wherein the counting element includes;

associating a potential response count attribute with the new message;

profiling the new message;

profiling the message responses; and

defining the potential response count for the new message based on a relatedness between the new message, a set of other new messages, and each of their respective actual response counts; and

wherein the re-transmitting element includes,

selecting the first sub-set of client computers based on the potential response count.

13. The method of claim 12 wherein the defining element includes;

calculating a “

new message relatedness”

between the new message and each new message within the set of other new messages;

retrieving the actual response counts for the other new messages; and

defining the potential response count as an average of the actual response counts respectively weighted by the “

new message relatedness”

of each of the other new messages.

14. The method of claim 13:

wherein the calculating element includes defining the “

new message relatedness”

as a distance between an expertise vector of the new message and an expertise vector of one of the other new messages.

15. The method of claim 12 wherein the defining element includes:

associating a potential response count (p) attribute with the new message;

profiling the new message;

profiling the message responses;

logically assigning a new message variable (m_new) to the new message;

logically assigning new message variables (m₁, m₂, . . . , m_n) respectively to a set of (n) other new messages;

calculating a “

new message relatedness”

(s( . . . )) between the new message (m_new) and each new message within a set of other new messages (m₁, m₂, . . . , m_n);

retrieving actual response counts (rc₁, rc₂, . . , rc_n) respectively corresponding to the other new messages; and

defining the potential response count (p) as equal to;

s(m_new, m_i)*rc_i

16. The method of claim 12 wherein the defining element includes:

associating a potential response count (p) attribute with the new message;

profiling the new message;

profiling the message responses;

logically assigning a new message variable (m_new) to the new message;

logically assigning new message variables (m₁, m₂, . . . , m_n) respectively to a set of (n) other new messages;

calculating a “

new message relatedness”

(s( . . . )) between the new message (m_new) and each new message within a set of other new messages (m₁, m₂, . . . , m_n);

retrieving actual response counts (rc₁, rc₂, . . . , rc_n) respectively corresponding to the other new messages; and

defining the potential response count (p) as equal to;

17. The method of claim 12, wherein tho selecting element includes:

defining the first sub-set as less than a predetermined number of client computers, if the potential response count is greater than a predetermined response count; and

defining the first sub-set as greater than the predetermined number of client computers, if the potential response count is less than the predetermined response count.

18. The method of claim 12, wherein the selecting element includes:

calculating a predicted message response quality level for the new message;

reducing a number of client computers in the first sub-set, if the predicted message response quality level is greater than a first predetermined quality level; and

increasing a number of client computers in the first sub-set, if the predicted message response quality level is less than the first predetermined quality level.

19. The method of claim 18, wherein:

the calculating a predicted message response quality level element includes, calculating a message response relatedness between each new message, within the set of other new messages, and corresponding message responses; and

the defining the potential response count includes, weighting the potential response count with the message response relatedness.

20. The method of claim 1, further comprising:

configuring the client computer as a server computer on the network.

21. The method of claim 20:

further comprising, requiring sending client computers to send only one copy of the new message to the server computer; and

wherein the re-transmitting element includes, selectively permitting client computers to pick-up new messages from the server.

22. A system for knowledge management, comprising:

means for generating a client profile on a client computer connected to a network;

means for storing the profile on the computer;

means for transmitting a new message;

means for scoring, at the client computer, the new message with respect to the client profile;

means for counting a number of message responses from client computers to the new message; and

means for re-transmitting the new message to a first sub-set of client computers on the network, based on the response count.

23. The system of claim 22, further comprising:

means for waiting for a predetermined time period;

means for retrieving an actual response count after the predetermined time period; and

means for re-transmitting the new message to a second sub-set of client computers on the network, if the actual response count remains less than a predetermined minimum response count.

24. The system of claim 22, further comprising:

means for configuring the client computer as a server computer on the network.

25. A method of software execution for knowledge management, comprising:

generating client profiles on client computers connected to a network;

storing the client profiles at the client computers;

transmitting a new message to a plurality of the client computers;

counting a number of responses made by the client computers to the new message; and

re-transmitting, based on the number of responses, the new message to a plurality of the client computers.

26. The method of claim 25 further comprising, waiting until expiration of a predetermined time period before re-transmitting the new message.

27. The method of claim 25 further comprising, deleting the new message from a plurality of the client computers once the number of responses reaches a predetermined number.

28. A method of software execution for knowledge management, comprising:

scoring a new message at a plurality of client computers;

if a score at a first client computer exceeds a threshold, then providing a client at the first client computer a time period to generate a response to the new message;

if the client at the first client computer does not generate the response within the time period, then automatically transmitting the new message from the first client computer to a first sub-set of the client computers;

counting a number of responses made by the client computers to the new message; and

re-transmitting the new message from the first client computer to a second sub-set of the client computers.

29. The method of claim 28, wherein the second sub-set is greater than and does not overlap the fist sub-set.

30. The method of claim 28 further comprising, if the number of responses exceeds a threshold, then ending re-transmission of the new message.