Populating a knowledgebase of an expert system

Patent metadata

Field	Value
Publication number	US-2016170976-A1
Application number	US-201615050586-A
Country	US
Kind code	A1
Filing date	Feb 23, 2016
Priority date	Jun 27, 2013
Publication date	Jun 16, 2016
Grant date	—

Abstract

A knowledgebase of an expert system is populated with rules inferred from a set of business processes that govern the manner in which the business interacts with users. Each business process contains an input, an output, an action, and a set of dependency relationships that relate pairs of the input, the output, and the action. Each process's input, output, action, and dependency relationships are translated, respectively, into a subject, an object, a predicate, and a set of dependency relationships among the subject, object, and predicate, of a natural-language rule. Each rule is stored in the expert system's knowledgebase as a directed graph, and nodes representing each stored subject, object, and predicate are assigned domain classifications as a function of characteristics of the business rule. These domain classifications are represented within the knowledgebase as a set of domain classifications determined as a further function of characteristics of the business rule.

First claim

What is claimed is:

1 . A method for populating a knowledgebase of an expert system with a set of rules stored as a directed graph, wherein each rule of the set of rules is stored as a subject node, an object node, a predicate node, and a set of dependencies that connect pairs of the subject node, the object node, and the predicate node, the method comprising:
a processor of a computer system representing a first process of a set of business processes as a first rule of the set of rules,
wherein the first process is associated with a natural-language interaction between the expert system and a user,
wherein the first rule is a rule for assigning meaning to the natural-language interaction as a function of the first business process,
wherein the first rule comprises a first subject, a first object, a first predicate, and a first set of rule relationships that each a identify a dependency relationship between two of the first subject, the first object, and the first predicate
wherein the first process comprises a first input, a first output, a first action, and a first set of process relationships that each a identify a dependency relationship between two of the first input, the first output, and the first action,
wherein the first subject is analogous to the first input, the first object is analogous to the first output, and the first predicate is analogous to the first action,
wherein the first set of rule relationships identify dependency relationships among the first subject, the first object, and the first predicate that are analogous to the dependency relationships among the first input, the first output, and the first action that are identified by the first set of process relationships,
wherein the representing the first process comprises adding three nodes to the directed graph that respectively represent the first subject, the first object, and the first predicate, and
wherein the representing the first process further comprises adding a set of dependencies to the directed graph that represent the first set of process relationships.

2 . The method of claim 1 , wherein the first rule is a rule for formulating a natural-language reply to a question comprised by the natural-language interaction.

3 . The method of claim 1 , wherein the first rule is a rule for formulating a natural-language question to the user in response to receiving the natural-language interaction.

4 . The method of claim 1 , wherein the first rule is a rule for inferring additional rules from the natural-language interaction.

5 . The method of claim 1 , wherein the first rule is a rule for invoking another rule of the set of rules stored in the knowledgebase in response to receiving the natural-language interaction.

6 . The method of claim 1 , wherein the knowledgebase is automatically populated as a function of information comprised by the set of business processes.

7 . The method of claim 1 ,
wherein a classification of a set of classifications is represented as a first classification node of the directed graph, wherein the set of classifications comprises a set of concept classifications and a set of activity classifications, wherein a first concept classification of the set of concept classifications associates an input domain with the first input and with the first subject, wherein a second concept classification of the set of concept classifications associates an output domain with the first output and with the first object, wherein a first activity classification of the set of activity classifications associates an action domain with the first action and with the first predicate, wherein a first classification relationship identifies a dependency relationship between a pair of classifications of the set of classifications, and where the first classification relationship is a function of a characteristic of the first business process, and wherein the representing the first process further comprises adding, as a further function of the characteristic, the first classification relationship to the directed graph as a first classification dependency of the set of dependencies.

8 . The method of claim 1 , further comprising providing at least one support service for at least one of creating, integrating, hosting, maintaining, and deploying computer-readable program code in the computer system, wherein the computer-readable program code in combination with the computer system is configured to implement the populating and the representing the first process.

9 . A computer program product, comprising a computer-readable hardware storage device having a computer-readable program code stored therein, said program code configured to be executed by a processor of a computer system to implement a method for populating a knowledgebase of an expert system with a set of rules stored as a directed graph, wherein each rule of the set of rules is stored as a subject node, an object node, a predicate node, and a set of dependencies that connect pairs of the subject node, the object node, and the predicate node, the method comprising:
the processor representing a first process of a set of business processes as a first rule of the set of rules,
wherein the first process is associated with a natural-language interaction between the expert system and a user,
wherein the first rule is a rule for assigning meaning to the natural-language interaction as a function of the first business process,
wherein the first rule comprises a first subject, a first object, a first predicate, and a first set of rule relationships that each a identify a dependency relationship between two of the first subject, the first object, and the first predicate
wherein the first process comprises a first input, a first output, a first action, and a first set of process relationships that each a identify a dependency relationship between two of the first input, the first output, and the first action,
wherein the first subject is analogous to the first input, the first object is analogous to the first output, and the first predicate is analogous to the first action,
wherein the first set of rule relationships identify dependency relationships among the first subject, the first object, and the first predicate that are analogous to the dependency relationships among the first input, the first output, and the first action that are identified by the first set of process relationships,
wherein the representing the first process comprises adding three nodes to the directed graph that respectively represent the first subject, the first object, and the first predicate, and
wherein the representing the first process further comprises adding a set of dependencies to the directed graph that represent the first set of process relationships.

10 . The computer program product of claim 9 , wherein the first rule is a rule for formulating a natural-language reply to a question comprised by the natural-language interaction.

11 . The computer program product of claim 9 , wherein the first rule is a rule for formulating a natural-language question to the user in response to receiving the natural-language interaction.

12 . The computer program product of claim 9 , wherein the first rule is a rule for inferring additional rules from the natural-language interaction.

13 . The computer program product of claim 9 , wherein the first rule is a rule for invoking another rule of the set of rules stored in the knowledgebase in response to receiving the natural-language interaction.

14 . The computer program product of claim 9 , wherein the knowledgebase is automatically populated as a fu

Assignees

• IBM

Inventors

• Byron Donna K
• Katahira Reinaldo T
• Krishnamurthy Lakshminarayanan
• Trim Craig M

Classifications

• G06F40/55 Primary

  Rule-based translation · CPC title

• G06N5/02 Primary

  Knowledge representation; Symbolic representation · CPC title

• G06N5/046

  Forward inferencing; Production systems · CPC title

• G06F17/2872 Primary

  Physics · mapped topic