Information evaluation

• US 5,799,304 A
• Filed: 10/23/1997
• Issued: 08/25/1998
• Est. Priority Date: 01/03/1995
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Miller, John David
• Primary Examiner(s): Black, Thomas G.
• Assistant Examiner(s): LEWIS, CHERYL RENEA
• Application Number:
  US08/956,091
• Time in Patent Office: 306 Days
• Field of Search: 707/3, 707/7, 707/201, 707/500
• US Class Current: 1/1
• CPC Class Codes: G06F 16/337 Profile generation, learning or modification
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching
• CPC Class Codes: Y10S 707/99937 Sorting
• CPC Class Codes: Y10S 707/99952 Coherency, e.g. same view to multiple users

Abstract

An automatic method and apparatus for processing information. Features in the information are recognized and occurrences of user-relevant features in the features of the information are determined. Based upon the occurrences of user-relevant features, evaluation values for portions of the information are determined based on a sum of the occurrences of the user-relevant features in each of the portions of the information. The evaluation values are then used to process the portions of the information. The processing may include an automatic recording process, squelch control, sorting process, display by evaluation value or other mechanisms.

Claims

1. A method for evaluating any computer-readable articles, the method comprising the computer-implemented steps of:

presenting any such articles to a set of agents for each article within a session, each agent including an expression, a voting bias, and an indication of the statistical reliability of the agent;

determining whether the expression for each agent matches the article;

determining whether the agent is sufficiently reliably to vote when the agent'"'"'s expression matches the article;

voting for or against the article according to the agent'"'"'s voting bias, if the agent is sufficiently reliable to vote; and

deriving an evaluation value for the article from a count of agents voting for the article and a count of agents voting against the article.

2. The method as claimed in claim 1 further including the step of:

receiving feedback which modifies how future articles should be evaluated.

3. The method as claimed in claim 2 further including the steps of:

updating, based on the feedback, the statistical reliability indication of each agent that was active during the session; and

determining, based on the feedback, whether to invert the voting bias of each agent that was active during the session.

4. The method as claimed in claim 2 further including the step of:

creating a new agent for a feature previously not having an agent.

5. The method as claimed in claim 1 wherein said voting bias is represented by a binary value.

6. The method as claimed in claim 1 further including the step of:

filtering out information which does not reach a certain threshold value.

7. The method as claimed in claim 1 further including the step of:

sorting information by accumulating values generated by a set of agents.

8. The method as claimed in claim 1 wherein the indication of statistical reliability of the agent is based on whether the agent had previously voted correctly within a standard deviation of the average number of correct votes for all agents.

9. An apparatus for evaluating articles, the apparatus comprising:

means for presenting any computer-readable articles to a set of agents for each article within a session, each agent including an expression, a voting bias, and an indication of the statistical reliability of the agent;

means for determining whether the expression for each agent matches the article;

means for determining whether the agent is sufficiently reliable to vote when the agent'"'"'s expression matches the article;

means for voting for or against the article according to the agent'"'"'s voting bias, if the agent is sufficiently reliable to vote; and

means for deriving an evaluation value for the article from a count of agents voting for the article and a count of agents voting against the article.

10. The apparatus as claimed in claim 9 further including:

means for receiving feedback which modifies how future articles should be evaluated.

11. The apparatus as claimed in claim 10 further including:

means for updating, based on the feedback, the statistical reliability indication of each agent that was active during the session; and

means for determining, based on the feedback, whether to invert the voting bias of each agent that was active during the session.

12. The apparatus as claimed in claim 10 further including:

means for creating a new agent for a feature previously not having an agent.

13. The apparatus as claimed in claim 9 wherein said voting bias is represented by a binary value.

14. The apparatus as claimed in claim 9 further including:

means for filtering out information which does not reach a certain threshold value.

15. The apparatus as claimed in claim 9 further including:

means for sorting information by the generated value.

16. The apparatus as claimed in claim 9 wherein the indication of statistical reliability of the agent is based on whether the agent had previously voted correctly within a standard deviation of the average number of correct votes for all agents.