Efficient service rules creation through subjective logic and temporal pattern recognition

• US 8,538,908 B2
• Filed: 01/26/2011
• Issued: 09/17/2013
• Est. Priority Date: 01/26/2011
• Status: Expired due to Fees

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Li, Zhiguo, Minhas, Rajinderjeet Singh
• Primary Examiner(s): Gaffin, Jeffrey A
• Assistant Examiner(s): FINK, THOMAS ANDREW
• Application Number:
  US13/014,019
• Publication Number:
  US 20120191638A1
• Time in Patent Office: 965 Days
• Field of Search: None
• US Class Current: 706/52
• CPC Class Codes: G06N 5/025 Extracting rules from data

Abstract

A method for generating service rules corresponding to business data is disclosed. A plurality of business related data is gathered from various sources. The data is combined using a subjective logic technique. The data is then evaluated for temporal patterns. Finally a set of service rules corresponding to the combined business data are developed.

Claims

1. A method for developing service rules, said method comprising:

gathering a plurality of business data sets comprising at least two of a textual description of a problem provided to a support center, a subjective logic opinion and associated problem code provided by a service technician, and error data provided by a machine;

determining a belief vector and uncertainty mass for each of said plurality of business data sets;

evaluating the combination of each belief vector and each uncertainty mass associated with each of said business data sets with a subjective logic technique;

producing a level of confidence based on said evaluation of the combination of each belief vector and each uncertainty mass associated with each of said business data sets with a subjective logic technique, wherein said level of confidence is indicative of a confidence in the reliability of said plurality of business data sets;

evaluating said combined business data for temporal patterns therein; and

generating a service rule corresponding to said produced level of confidence in said business data and said temporal error patterns.

2. The method of claim 1 wherein evaluating said combined business data for temporal patterns therein, further comprises:

determining an event sequence;

defining a window associated with said event sequence;

evaluating said event sequence for a failure signature; and

determining the frequency said window covers said failure signature.

3. The method of claim 2 wherein said failure signature comprises at least one of the following types of failure signatures:

a serial failure signature;

a parallel failure signature;

or

a composite failure signature.

4. The method of claim 2 further comprising utilizing said failure signature to generate said service rule to predict an occurrence of a failure event.

5. A method for developing service rules comprising:

gathering a plurality of business data sets comprising machine service data from a plurality of sources at least two of a textual description of a problem provided to a support center, a subjective logic opinion and associated problem code provided by a service technician, and error data provided by a machine;

combining said business data sources;

producing a level of confidence in said combined business data wherein said level of confidence is indicative of a confidence in the reliability of said plurality of business data sets by;

determining a belief vector and an uncertainty mass for each of said business data sets; and

evaluating the combination of each belief vector and each uncertainty mass associated with each of said business data sets with a subjective logic technique;

evaluating said combined business data for temporal patterns by;

determining an event sequence;

defining a window associated with said event sequence;

evaluating said event sequence for a failure signature; and

determining the frequency said window covers said failure signature; and

generating a service rule corresponding to said combined business data and said temporal error patterns.

6. The method of claim 5 wherein said failure signature comprises at least one of the following types of failure signatures:

a serial failure signature;

a parallel failure signature;

or

a composite failure signature.

7. The method of claim 5 further comprising utilizing said failure signature to generate said plurality of service rules to predict the occurrence of a failure event.

8. The method of claim 5 wherein evaluating each of said business data set'"'"'s belief vector and uncertainty mass with said subjective logic technique produces an estimate of a system code and a problem code.

9. A system for developing service rules, comprising:

a processor;

a data bus coupled to said processor; and

a computer-usable medium embodying computer code, said computer-usable medium being coupled to said data bus, said computer program code comprising instructions executable by said processor and configured for;

gathering a plurality of business data sets comprising at least two of a textual description of a problem provided to a support center, a subjective logic opinion and associated problem code provided by a service technician, and error data provided by a machine;

determining a belief vector and uncertainty mass for each of said plurality of business data sets;

evaluating the combination of each belief vector and each uncertainty mass associated with each of said business data sets with a subjective logic technique;

producing a level of confidence based on said evaluation of the combination of each belief vector and each uncertainty mass associated with each of said business data sets with a subjective logic technique, wherein said level of confidence is indicative of a confidence in the reliability of said plurality of business data sets;

evaluating said combined business data for temporal patterns therein; and

generating a service rule corresponding to said produced level of confidence in said business data and said temporal error patterns.

10. The system of claim 9 wherein said instructions are further configured for:

determining an event sequence;

defining a window associated with said event sequence;

evaluating said event sequence for a failure signature; and

determining the frequency said window covers said failure signature.

11. The system of claim 10 wherein said failure signature comprises at least one of the following types of failure signatures:

a serial failure signature;

a parallel failure signature;

or

a composite failure signature.

12. The system of claim 10 wherein said failure signature is utilized to generate a plurality of service rules to predict the occurrence of a failure event.