Efficient rule execution in decision services

• US 9,798,973 B2
• Filed: 08/16/2013
• Issued: 10/24/2017
• Est. Priority Date: 08/23/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Baudel, Thomas, Feillet, Pierre D.
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Nilsson, Eric
• Application Number:
  US13/969,135
• Publication Number:
  US 20140058995A1
• Time in Patent Office: 1,530 Days
• Field of Search: None
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/025 Extracting rules from data

Abstract

A method, system, and/or computer program product performs rule execution in decision services. One or more specialized rulesets classified by decision partitions are based on defined clustering predicates of decision service signatures for business rules. A decision request is received from a client, and one or more parameters of the decision request are input to a clustering predicate based on decision service signatures to produce a decision category. A specialized ruleset for the decision request is determined based on the decision category, and the decision request is forwarded for processing to the specialized ruleset. A decision input is then returned from the specialized ruleset.

Claims

1. A method for rule execution in decision services, comprising:

providing, by one or more processors, one or more specialized rulesets classified by decision partitions based on defined clustering predicates of decision service signatures for business rules, wherein a clustering predicate is a function that determines a decision category based on parameters that are input into the function, wherein the clustering predicate is based on the decision service signature, wherein the decision service signatures define input parameters and output parameters for a decision service, wherein the decision service is an autonomous component that receives decision requests and return decision outputs;

receiving, by one or more processors, a decision request from a client, wherein the decision service redirects the decision request to a particular set of business rules;

inputting, by one or more processors, one or more parameters of the decision request to a clustering predicate based on the decision service signatures to produce the decision category;

determining, by one or more processors, a specialized ruleset for the decision request based on the decision category;

forwarding, by one or more processors, the decision request for processing by the specialized ruleset;

returning, by one or more processors, a decision output from the specialized ruleset;

defining, by one or more processors, the clustering predicate at business rules authoring time;

analyzing, by one or more processors, business rules to identify decision partitions based on the clustering predicate and to classify the business rules per decision partition to form specialized rulesets, wherein said analyzing the business rules includes;

scanning, by one or more processors, all business rules for all conditions in the business rules;

scanning, by one or more processors, all conditions for all tests performed on the decision partitions;

adding, by one or more processors, a new decision partition;

adding a test to the new decision partition; and

adding the test to a default decision partition.

2. The method as claimed in claim 1, further comprising:

providing a default ruleset;

determining, by one or more processors, if no specialized ruleset for the decision request is found based on the parameters of the decision request;

forwarding, by one or more processors, the decision request for processing by the default ruleset;

returning, by one or more processors, a decision output from the default ruleset.

3. The method as claimed in claim 2, wherein the default ruleset is a full ruleset from which the specialized rulesets have been partitioned.

4. The method as claimed in claim 1, further comprising:

manually extending decision partitions for specialized rulesets.

5. The method as claimed in claim 1, further comprising:

simplifying, by one or more processors, business rules by partition by factorizing predicate tests that apply to the partition.

6. The method as claimed in claim 1, wherein analyzing business rules to identify decision partitions includes:

for all partitions including a default, and for all business rules with a test in a partition;

subtracting, by one or more processors, the test from the business rule;

adding, by one or more processors, the business rule to the specialized ruleset.

7. The method as claimed in claim 1, wherein analyzing business rules to identify decision partitions includes:

removing, by one or more processors, rules that test against another value.

8. The method as claimed in claim 1, wherein analyzing business rules to identify decision partitions includes:

reordering, by one or more processors, tests, conditions and rules.

9. The method as claimed in claim 1, wherein providing one or more specialized rulesets classified by decision partitions based on clustering predicates for business rules includes:

partitioning, by one or more processors, the one or more specialized rulesets by decomposing a monolithic multi-purpose ruleset.

10. A system for rule execution in decision services, the system comprising:

a processor, a computer readable memory, and a computer readable storage media;

first program instructions to provide one or more specialized rulesets classified by decision partitions based on defined clustering predicates of decision service signatures for business rules, wherein a clustering predicate is a function that determines a decision category based on parameters that are input into the function, wherein the clustering predicate is based on the decision service signature, wherein the decision service signatures define input parameters and output parameters for a decision service, wherein the decision service is an autonomous component that receives decision requests and return decision outputs;

second program instructions to receive a decision request from a client, wherein the decision service redirects the decision request to a particular set of business rules;

third program instructions to input one or more parameters of the decision request to a clustering predicate based on decision service signatures to produce the decision category;

fourth program instructions to determine a specialized ruleset for the decision request based on the decision category;

fifth program instructions to forward the decision request for processing by the specialized ruleset;

sixth program instructions to return a decision output from the specialized ruleset;

seventh program instructions to define the clustering predicate at business rules authoring time;

eighth program instructions to analyze business rules to identify decision partitions based on the clustering predicate and to classify the business rules per decision partition to form specialized rulesets, wherein said analyzing the business rules includes;

scanning all business rules for all conditions in the business rules;

scanning all conditions for all tests performed on the decision partitions;

adding a new decision partition;

adding a test to the new decision partition; and

adding the test to a default decision partition; and

wherein

said first, second, third, fourth, fifth, sixth, seventh and eighth program instructions are stored on said computer readable storage media for execution by said processor via said computer readable memory.

11. The system as claimed in claim 10, further comprising:

ninth program instructions to provide a default ruleset;

tenth program instructions to determine if no specialized ruleset for the decision request is found based on the parameters of the decision request;

eleventh program instructions to forward the decision request for processing by the default ruleset; and

twelfth program instructions to return a decision output from the default ruleset; and

wherein

said ninth, tenth, eleventh, and twelfth program instructions are stored on said computer readable storage media for execution by said processor via said computer readable memory.

12. The system as claimed in claim 11, wherein the default ruleset is a full ruleset from which the specialized rulesets have been partitioned.

13. A computer program product for rule execution in decision services, the computer program product comprising a non-transitory computer readable storage medium having computer-readable program code embodied therewith, the computer-readable program code configured to:

provide one or more specialized rulesets classified by decision partitions based on defined clustering predicates of decision service signatures for business rules, wherein a clustering predicate is a function that determines a decision category based on parameters that are input into the function, wherein the clustering predicate is based on the decision service signature, wherein the decision service signatures define input parameters and output parameters for a decision service, wherein the decision service is an autonomous component that receives decision requests and return decision outputs;

receive a decision request from a client, wherein the decision service redirects the decision request to a particular set of business rules;

input one or more parameters of the decision request to a clustering predicate based on decision service signatures to produce the decision category;

determine a specialized ruleset for the decision request based on the decision category;

forward the decision request for processing by the specialized ruleset;

return a decision output from the specialized ruleset;

define the clustering predicate at business rules authoring time;

analyze business rules to identify decision partitions based on the clustering predicate and to classify the business rules per decision partition to form specialized rulesets, wherein said analyzing the business rules includes;

scanning all business rules for all conditions in the business rules;

scanning all conditions for all tests performed on the decision partitions;

adding a new decision partition;

adding a test to the new decision partition; and

adding the test to a default decision partition.

14. The computer program product of claim 13, wherein the computer-readable program code is further configured to:

provide a default ruleset;

determine if no specialized ruleset for the decision request is found based on the parameters of the decision request;

forward the decision request for processing by the default ruleset;

return a decision output from the default ruleset.

15. The computer program product of claim 14, wherein the default ruleset is a full ruleset from which the specialized rulesets have been partitioned.

16. The computer program product of claim 13, wherein the computer-readable program code is further configured to:

simplify business rules by partition by factorizing predicate tests that apply to the partition.