Dialogue control system, method and computer readable storage medium, and multidimensional ontology processing system, method and computer readable storage medium

• US 8,694,453 B2
• Filed: 12/15/2009
• Issued: 04/08/2014
• Est. Priority Date: 04/30/2009
• Status: Active Grant

• Current Assignee: OKI Electric Industry Company Limited
• Original Assignee: OKI Electric Industry Company Limited
• Inventors: Kitamura, Mihoko, Murata, Toshiki, Sukehiro, Tatsuya, Shimohata, Sayori, Orihara, Ikuo, Ikeno, Atsushi, Kato, Shigehiro, Tanaka, Miho
• Primary Examiner(s): VINCENT, DAVID ROBERT
• Application Number:
  US13/145,516
• Publication Number:
  US 20110282913A1
• Time in Patent Office: 1,575 Days
• Field of Search: 706/20, 706/45, 706/12
• US Class Current: 706/45
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/367 Ontology

Abstract

A dialogue control system that uses an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts to determine questions for a user and implements the questions, and implements a dialogue in accordance with received user answers. One or more dimension-classified ontologies and a basic ontology are stored, a concept from a user response is specified as a condition, and one or a plural number of the dimension-classified ontologies that meet the specified condition are fetched. The content of the fetched dimension-classified ontology/ies is overwritten on the basic ontology, in order according to dimension priorities from the dimension-classified ontology with the lowest dimension priority, and a multidimensional ontology is generated. The dialogue control system uses this multidimensional ontology to control the dialogue.

Claims

1. An ontology processing system that performs predetermined processing using an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts, the ontology processing system comprising:

a dimension-classified ontology storage unit that stores at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

a basic ontology storage unit that stores a basic ontology including domain knowledge content for when there is no condition;

a dimension priority storage unit that stores dimension priorities representing priorities of the respective dimensions;

a dimension-classified ontology fetcher that, when a condition is specified, fetches from the dimension-classified ontology storage unit one or more dimension-classified ontologies that meet the specified condition;

a multidimensional ontology generator that overwrites the content of the fetched dimension-classified ontologies onto the basic ontology according to the dimension priorities of the dimension priority storage unit, in order from the dimension-classified ontology with the lowest dimension priority, and generates a multidimensional ontology to be used with the specified condition; and

a processer that performs predetermined processing using the multidimensional ontology generated by the multidimensional ontology generator.

2. The ontology processing system according to claim 1, wherein

the domain knowledge includes concepts and relationships between concepts, and dialogue progress information that specifies movement relationships from concepts employed in questions to concepts to be employed in subsequent questions, and

the predetermined processing includes determining questions for a user and implementing the questions, receiving user answers to the questions, deciding on concepts to be employed in subsequent questions on the basis of the multidimensional ontologies generated by the multidimensional ontology generator in accordance with conditions according to the user answers, and providing the questions to the user.

3. A non-transitory computer readable storage medium storing an ontology processing program that performs predetermined processing using an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts, the ontology processing comprising:

storing, in a dimension-classified ontology storage unit, at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

storing, in a basic ontology storage unit, a basic ontology including domain knowledge content for when there is no condition;

storing, in a dimension priority storage unit, dimension priorities representing priorities of the respective dimensions;

when a condition is specified, fetching from the at least one stored dimension-classified ontologies one or more dimension-classified ontologies that meet the specified condition;

generating a multidimensional ontology to be used with the specified condition, including overwriting the content of the fetched dimension-classified ontologies onto the basic ontology according to the stored dimension priorities, in order from the dimension-classified ontology with the lowest dimension priority; and

performing predetermined processing using the generated multidimensional ontology.

4. A dialogue control system that uses an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts to be used in questions, determines questions for a user and implements the questions, receives user answers to the questions, and controls a dialogue in accordance with the user answers,

wherein the domain knowledge includes the concepts and relationships between concepts, and dialogue progress information that specifies movement relationships from concepts employed in questions to concepts to be employed in subsequent questions, and

the dialogue control system includes:

a dimension-classified ontology storage unit that stores at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

a basic ontology storage unit that stores a basic ontology including domain knowledge content for when there is no condition;

a dimension priority storage unit that stores dimension priorities representing priorities of the respective dimensions;

a dimension-classified ontology fetcher that, when a concept in the user answers is specified as a condition, fetches from the dimension-classified ontology storage unit one or more dimension-classified ontologies that meet the specified condition;

a multidimensional ontology generator that overwrites the content of the fetched dimension-classified ontologies onto the basic ontology according to the dimension priorities of the dimension priority storage unit, in order from the dimension-classified ontology with the lowest dimension priority, and generates a multidimensional ontology to be used with the condition; and

a dialogue controller that decides on a concept to be employed in a subsequent question on the basis of the dialogue progress information of the multidimensional ontology generated by the multidimensional ontology generator, and provides the question to the user.

5. The dialogue control system according to claim 4, wherein

the domain knowledge specifies question texts for relationships between concepts, and

the dialogue controller outputs the question text specified for the concept to be employed in the subsequent question to the user.

6. The dialogue control system according to claim 4, wherein

the domain knowledge specifies at least check texts and link texts for relationships between concepts,

the multidimensional ontology generator dynamically changes the check texts and link texts by generating multidimensional ontologies with concepts in the user answers as conditions, and

the dialogue controller outputs the dynamically changed check texts and link texts to the user.

7. The dialogue control system according to claim 4, wherein

the basic ontology and the dimension-classified ontologies include inference definition information that infers concepts from other concepts,

the multidimensional ontology generator dynamically changes the inference definitions by generating multidimensional ontologies with concepts in the user answers serving as conditions, and

the dialogue controller infers concepts that correspond with concepts in the user answers on the basis of the dynamically changed inference definitions.

8. The dialogue control system according to claim 4, further comprising, preceding the dialogue controller, an intention analyser that performs analysis of intentions of the user from concepts in the user answers received in response to the questions implemented for the user.

9. The dialogue control system according to claim 8 wherein, if the intention analyser detects at least a predetermined number of repetitions of a particular expression or term on the basis of concepts in the user answers, the multidimensional ontology generator generates a multidimensional ontology with the particular expression or term serving as a condition.

10. A non-transitory computer readable storage medium storing a dialogue control program that uses an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts to be used in questions, determines questions for a user and implements the questions, receives user answers to the questions, and controls a dialogue in accordance with the user answers,

wherein the domain knowledge includes the concepts and relationships between concepts, and dialogue progress information that specifies movement relationships from concepts employed in questions to concepts to be employed in subsequent questions, and

the dialogue control includes:

storing, in a dimension-classified ontology storage unit, at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

storing, in a basic ontology storage unit, a basic ontology including domain knowledge content for when there is no condition;

storing, in a dimension priority storage unit, dimension priorities representing priorities of the respective dimensions;

when a concept is specified as a condition in the user answers, fetching from the at least one stored dimension-classified ontologies one or more dimension-classified ontologies that meet the specified condition;

generating a multidimensional ontology to be used with the condition, including overwriting the content of the fetched dimension-classified ontologies onto the basic ontology according to the stored dimension priorities, in order from the dimension-classified ontology with the lowest dimension priority; and

deciding on a concept to be employed in a subsequent question on the basis of the dialogue progress information of the generated multidimensional ontology, and providing the question to the user.

11. An ontology processing method that performs predetermined processing using an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts, the ontology processing method comprising:

storing, in a dimension-classified ontology storage unit, at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

storing, in a basic ontology storage unit, a basic ontology including domain knowledge content for when there is no condition;

storing, in a dimension priority storage unit, dimension priorities representing priorities of the respective dimensions;

when a condition is specified, fetching from the at least one stored dimension-classified ontologies one or more dimension-classified ontologies that meet the specified condition;

generating a multidimensional ontology to be used with the specified condition, including overwriting the content of the fetched dimension-classified ontologies onto the stored basic ontology according to the stored dimension priorities, in order from the dimension-classified ontology with the lowest dimension priority; and

performing predetermined processing using the generated multidimensional ontology.

12. A dialogue control processing method that uses an ontology that systematically expresses domain knowledge representing concepts and relationships between concepts to be used in questions, determines questions for a user and implements the questions, receives user answers to the questions, and controls a dialogue in accordance with the user answers,

wherein the domain knowledge includes the concepts and relationships between concepts, and dialogue progress information that specifies movement relationships from concepts employed in questions to concepts to be employed in subsequent questions, and

the dialogue control includes:

storing, in a dimension-classified ontology storage unit, at least one dimension-classified ontology, for each of dimensions for combinations of conditions, the dimension-classified ontologies differing in domain knowledge content;

storing, in a basic ontology storage unit, a basic ontology including domain knowledge content for when there is no condition;

storing, in a dimension priority storage unit, dimension priorities representing priorities of the respective dimensions;

when a concept is specified as a condition in the user answers, fetching from the at least one stored dimension-classified ontologies one or more dimension-classified ontologies that meet the specified condition;

generating a multidimensional ontology to be used with the condition, including overwriting the content of the fetched dimension-classified ontologies onto the basic ontology according to the stored dimension priorities, in order from the dimension-classified ontology with the lowest dimension priority; and

deciding on a concept to be employed in a subsequent question on the basis of the dialogue progress information of the generated multidimensional ontology, and providing the question to the user.