Statistical methods and apparatus for records management

• US 7,451,155 B2
• Filed: 10/05/2005
• Issued: 11/11/2008
• Est. Priority Date: 10/05/2005
• Status: Active Grant

• Current Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Original Assignee: AT&T Intellectual Property I LP (AT&T, Inc.)
• Inventors: Apple, William, Slackman, Richard, O'Brien, Don
• Primary Examiner(s): Wu, Yicun
• Application Number:
  US11/245,365
• Publication Number:
  US 20070088715A1
• Time in Patent Office: 1,133 Days
• Field of Search: 707/100
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/10 Office automation; Time management

Abstract

Statistical methods and apparatus for records management are disclosed. An example method for records management disclosed herein comprises classifying a record into at least one of a plurality of categories based on a statistical value, and performing an operation with respect to the record according to a rule associated with the at least one of the plurality of categories, wherein the operation is performed based on the statistical value.

Claims

1. A computer implemented method for records management comprising:

performing a first assessment having a first accuracy and a separate second assessment having a second accuracy to determine respective first and second statistical values, each of the first and second statistical values representing an association between a record and a first category in a plurality of categories;

combining the first statistical value weighted by the first accuracy and the second statistical value weighted by the second accuracy to determine a composite statistical value representing the association between the record and the first category;

classifying the record into the first category when the composite statistical value at least meets a first threshold, wherein the composite statistical value is determined to classify the record;

retrieving a first rule associated with the first category after the record is classified in the first category; and

evaluating the composite statistical value against a second threshold associated with the first rule to determine whether to perform an operation with respect to the record according to the first rule associated with the first category.

2. A method as defined in claim 1 wherein the composite statistical value comprises at least one of a confidence level or a probability.

3. A method as defined in claim 1 wherein performing the first assessment comprises determining a likelihood that the record belongs to the first category in the plurality of categories, and wherein the first accuracy is at least one of determined theoretically or determined experimentally based on evaluation of previous classification assessments.

4. A method as defined in claim 3 wherein the likelihood is based on at least one of an automated assessment or a manual assessment.

5. A method as defined in claim 3 wherein the likelihood comprises at least one of a numeric value or a linguistic value.

6. A method as defined in claim 3 wherein determining the first statistical value comprises converting the likelihood from a linguistic value to a numeric value.

7. A method as defined in claim 1 wherein the operation comprises at least one of producing the record, retaining the record, destroying the record, issuing an instruction to produce the record, issuing an instruction to retain the record or issuing an instruction to destroy the record.

8. A method as defined in claim 1 wherein performing the operation with respect to the record according to the first rule associated with the first category comprises performing the operation if the composite statistical value is not less than the second threshold.

9. A method as defined in claim 1 wherein performing the operation with respect to the record according to the first rule associated with the first category comprises performing the operation if the composite statistical value is a maximum value in a set of statistical values associated with a set of categories into which the record is classified.

10. A method as defined in claim 9 wherein the set of statistical values is a first set of statistical values determined by combining at least two statistical values in a second set of statistical values associated with the set of categories, wherein the two statistical values in the second set of statistical values correspond respectively to two categories associated with at least one substantially similar rule.

11. A method as defined in claim 9 wherein performing the operation with respect to the record according to the first rule associated with the first category further comprises performing the operation based on a property of the first rule if the composite statistical value is one of multiple statistical values substantially equal to the maximum value.

12. A method as defined in claim 11 wherein the property is associated with a period of time.

13. A method as defined in claim 1 further comprising electronically storing at least one of the record or a reference to a location of the record.

14. A method as defined in claim 13 further comprising storing the composite statistical value such that the composite statistical value is associated with the first category and at least one of the record or the reference to the location of the record.

15. A method as defined in claim 1 wherein the record comprises at least one of a medical record, an accounting record, a legal record, an insurance record, an employment record, a payroll record or a shipping record.

16. A method as defined in claim 1 wherein the first rule comprises at least one of a retention rule, a confidentiality rule or a security rule.

17. An article of manufacture storing machine readable instructions which, when executed, cause a machine to:

perform a first assessment having a first accuracy and a separate second assessment having a second accuracy to determine respective first and second statistical values, each of the first and second statistical values representing an association between a record and a first category in a plurality of categories;

combine the first statistical value weighted by the first accuracy and the second statistical value weighted by the second accuracy to determine a composite statistical value representing the association between the record and the first category;

classify the record into the first category when the composite statistical value at least meets a first threshold, wherein the composite statistical value is determined to classify the record;

retrieve a first rule associated with the first category after the record is classified in the first category; and

evaluate the composite statistical value against a second threshold associated with the first rule to determine whether to perform an operation with respect to the record according to the first rule associated with the first category.

18. An article of manufacture as defined in claim 17 wherein the composite statistical value comprises at least one of a confidence level or a probability.

19. An article of manufacture as defined in claim 17 wherein the machine readable instructions, when executed, cause the machine to perform the first assessment by determining a likelihood that the record belongs to the first category in the plurality of categories, and wherein the first accuracy is at least one of determined theoretically or determined experimentally based on evaluation of previous classification assessments.

20. An article of manufacture as defined in claim 17 wherein the operation comprises at least one of producing the record, retaining the record, destroying the record, issuing an instruction to produce the record, issuing an instruction to retain the record or issuing an instruction to destroy the record.

21. An article of manufacture as defined in claim 17 wherein the machine readable instructions, when executed, further cause the machine to electronically store at least one of the record or a reference to a location of the record, and to store the composite statistical value such that the composite statistical value is associated with the at least one of the record or the reference to the location of the record.

22. A system comprising computer executed operation for records management comprising

a classifier configured to classify a record into a first category in a plurality of categories based on a first statistical value relating the record to the first category and to classify the record into a second category in the plurality of categories based on a second statistical value relating the record to the second category, wherein the classifier is further configured to determine the first statistical value and the second statistical value to classify the record;

and

a record processing unit configured (i) to combine the first statistical value and the second statistical value after the record has been classified to form a composite statistical value when the first category and the second category have a common rule and (ii) to evaluate the composite statistical value to determine whether to perform an operation with respect to the record according to the common rule when either the first category or the second category has been designated for processing.

23. A system as defined in claim 22 wherein the composite statistical value comprises at least one of a confidence level or a probability.

24. A system as defined in claim 22 wherein the classifier is configured to determine a first likelihood that the record belongs to the first category and a second likelihood that the record belongs to the second category.

25. A system as defined in claim 22 wherein the record processing unit comprises at least one of a retention unit configured to determine whether to retain or destroy the record, or a production unit configured to at least one of produce the record or issue an instruction to produce the record.

26. A system as defined in claim 22 wherein the record processing unit is configured to perform the operation if the composite statistical value is not less than a threshold.

27. A system as defined in claim 22 wherein the record processing unit is configured to perform the operation based on a comparison of the composite statistical value to another statistical value associated with another one of the plurality of categories.

28. A system as defined in claim 22 further comprising a storage unit to store at least one of the record or a reference to a location of the record, and to store the first statistical value and the second statistical value such that the first statistical value is associated with the first category and at least one of the record or the reference to the location of the record and the second statistical value is associated with the second category and at least one of the record or the reference to the location of the record.

29. An apparatus comprising computer executed operation to process a record comprising:

a category map selector to process a category map associated with the record, wherein the category map comprises a category field associated with a category from a plurality of categories, a statistical value field associated with a statistical value representing a likelihood that the record belongs to the category, and a management rule field associated with at least one rule associated with an operation to be performed with respect to records belonging to the category;

a statistical value combiner to determine the statistical value associated with the statistical value field to classify the record into the category by combining a plurality of statistical values weighted by a respective plurality of assessment accuracies, the plurality of statistical values determined from a respective plurality of separate assessments performed to determine whether the record should be classified into the category;

a rule retriever to retrieve a management rule associated with the management rule field; and

a rule evaluator to evaluate the statistical value associated with the statistical value field to determine whether the likelihood of the record belonging to the category is sufficient to process the management rule.

30. An apparatus as defined in claim 29 wherein the-statistical value comprises at least one of a confidence level or a probability.

31. An apparatus as defined in claim 29 wherein the management rule comprises a retention period corresponding to a period of time to retain records in the category.

32. An apparatus as defined in claim 29 wherein the management rule field is structured to store the management rule or the management rule field is structured to store a reference to the management rule.

33. An apparatus as defined in claim 32 wherein the rule retriever is configured to retrieve the management rule from a repository of management rules based on the stored reference.

34. An apparatus as defined in claim 29 wherein the rule evaluator is configured to determine whether the management rule applies to the record based on the statistical value.

35. An apparatus as defined in claim 34 wherein the rule evaluator is configured to determine that the management rule applies to the record by comparing the statistical value to a predetermined criterion.

36. An apparatus as defined in claim 31 wherein the rule evaluator is configured to determine to destroy the record if the statistical value is not less than a predetermined threshold and the retention period has been exceeded.

37. An apparatus as defined in claim 29 further comprising a rule combiner to combine two or more management rules associated with two or more categories based on statistical values associated with the two or more management rules.

38. An apparatus as defined in claim 37 wherein the rule combiner is configured to combine the two or more management rules if the two or more management rules comprise substantially similar parameters.

39. An apparatus as defined in claim 38 wherein the substantially similar parameters comprise retention periods.

40. An apparatus comprising computer executed operation to produce a record comprising:

a category map selector to process a category map associated with the record, wherein the category map comprises a first category field and a second category field associated respectively with a first category and a second category from a plurality of categories, and a first statistical value field and a second statistical value field associated respectively with a first statistical value representing a first likelihood that the record belongs to the first category and a second statistical value representing a second likelihood that the record belongs to the second category, wherein the first and the second statistical values are determined to classify the record into the first and second categories;

a statistical value processor to process the first statistical value associated with the first statistical value field and the second statistical value associated with the second statistical value field to determine whether the likelihood of the record belonging to either the first category or the second category is sufficient to produce the record in response to a query to produce records associated with the first and second categories; and

a production evaluator to at least one of produce the record or issue an instruction to produce the record in response to the determination by the statistical value processor.

41. An apparatus as defined in claim 40 wherein the first statistical value comprises at least one of a confidence level or a probability.

42. An apparatus as defined in claim 40 wherein the statistical value processor is configured to compare the first statistical value and the second statistical value to a criterion.

43. An apparatus as defined in claim 42 wherein the criterion represents a required degree of association between the record and a particular category to produce the record in response to a query to produce records belonging to the particular category.

44. An apparatus as defined in claim 40 wherein the production evaluator is configured to at least one of produce the record or issue the instruction to produce the record if either the first statistical value or the second statistical value is not less than a threshold.