Ontology-based automatic bootstrapping of state-based dialog systems

• US 10,839,160 B2
• Filed: 01/19/2018
• Issued: 11/17/2020
• Est. Priority Date: 01/19/2018
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Sen, Jaydeep, Jain, Parag, Saha, Diptikalyan, Mittal, Ashish
• Primary Examiner(s): Armstrong, Angela A
• Application Number:
  US15/875,303
• Publication Number:
  US 20190228068A1
• Time in Patent Office: 1,033 Days
• CPC Class Codes: G06F 16/242 Query formulation
• CPC Class Codes: G06F 16/9024 Graphs; Linked lists G06F16/9027 takes precedence
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Methods, systems, and computer program products for bootstrapping of state-based dialog systems are provided herein. A computer-implemented method includes determining parameters for state automata by partitioning an ontology graph into sub-graphs and a knowledge graph into sub-graphs, wherein the ontology graph and the knowledge graph are based on a user question and domain knowledge pertaining to the user question; generating a structured query for each of the sub-graphs; determining intentions of a dialog pertaining to the at least one user question by translating each of the generated structured queries to a respective natural language query; creating one or more dialog states for each of the determined dialog intentions; creating one or more connecting dialog states between pairs of the created dialog states; and generating an automata dialog framework associated with the user question based on the created dialog states and the created connecting dialog states.

Claims

1. A computer-implemented method, the method comprising steps of:

determining one or more parameters for state automata for use by one or more automated conversation exchange programs by partitioning each of (i) at least one ontology graph into multiple sub-graphs and (ii) at least one knowledge graph into multiple sub-graphs, wherein each of (i) the at least one ontology graph and (ii) the at least one knowledge graph is based on both at least one user question and domain knowledge pertaining to the at least one user question, and wherein each state in the state automata represents a structured action comprising at least one of a query and a command;

generating a structured query for each of the multiple sub-graphs;

determining one or more intentions of a dialog pertaining to the at least one user question by translating each of the generated structured queries to a respective natural language query;

creating one or more dialog states for each of the determined dialog intentions;

creating one or more connecting dialog states between pairs of the created dialog states; and

generating an automata dialog framework associated with the at least one user question based on (i) the created dialog states and (ii) the created connecting dialog states;

wherein the steps are carried out by at least one computing device.

2. The computer-implemented method of claim 1, wherein said determining one or more parameters for state automata comprises creating one or more ontology-independent structured action classes.

3. The computer-implemented method of claim 2, wherein said determining one or more parameters for state automata comprises creating one or more ontology-dependent parameterized actions based on the one or more ontology-independent structured action classes.

4. The computer-implemented method of claim 3, wherein said determining one or more parameters for state automata comprises generating, for each of the ontology-dependent parameterized actions, an action intention.

5. The computer-implemented method of claim 3, wherein said determining one or more parameters for state automata comprises generating one or more concrete actions based on the one or more ontology-dependent parameterized actions.

6. The computer-implemented method of claim 3, wherein said determining one or more parameters for state automata comprises deriving an identification of one or more entities.

7. The computer-implemented method of claim 6, wherein said determining one or more parameters for state automata comprises generating one or more concrete actions, using values from the one or more entities, based on the one or more ontology-dependent parameterized actions.

8. The computer-implemented method of claim 5, wherein said determining one or more parameters for state automata comprises generating one or more natural language queries corresponding to each of the concrete actions.

9. The computer-implemented method of claim 4, wherein said creating one or more dialog states comprises creating one or more action dialog states for each of the generated action intentions.

10. The computer-implemented method of claim 9, wherein said creating one or more dialog states comprises representing a difference between two of the action dialog states using one or more natural language actions.

11. The computer-implemented method of claim 10, comprising:

creating a non-sequential utterance intention pertaining to the difference.

12. The computer-implemented method of claim 11, wherein said creating one or more dialog states comprises creating a transition dialog state associated with the two action dialog states.

13. The computer-implemented method of claim 12, wherein said generating an automata dialog framework comprises creating a dialog flow comprising the two action dialog states and the transition dialog state.

14. The computer-implemented method of claim 12, wherein said generating an automata dialog framework comprises assigning the created non-sequential utterance intention to the transition dialog state.

15. A computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computing device to cause the computing device to:

determine one or more parameters for state automata for use by one or more automated conversation exchange programs by partitioning each of (i) at least one ontology graph into multiple sub-graphs and (ii) at least one knowledge graph into multiple sub-graphs, wherein each of (i) the at least one ontology graph and (ii) the at least one knowledge graph is based on both at least one user question and domain knowledge pertaining to the at least one user question, and wherein each state in the state automata represents a structured action comprising at least one of a query and a command;

generate a structured query for each of the multiple sub-graphs;

determine one or more intentions of a dialog pertaining to the at least one user question by translating each of the generated structured queries to a respective natural language query;

create one or more dialog states for each of the determined dialog intentions;

create one or more connecting dialog states between pairs of the created dialog states; and

generate an automata dialog framework associated with the at least one user question based on (i) the created dialog states and (ii) the created connecting dialog states.

16. The computer program product of claim 15, wherein said determining one or more parameters for state automata comprises creating one or more ontology-independent structured action classes.

17. The computer program product of claim 16, wherein said determining one or more parameters for state automata comprises creating one or more ontology-dependent parameterized actions based on the one or more ontology-independent structured action classes.

18. A system comprising:

a memory; and

at least one processor operably coupled to the memory and configured for;

determining one or more parameters for state automata for use by one or more automated conversation exchange programs by partitioning each of (i) at least one ontology graph into multiple sub-graphs and (ii) at least one knowledge graph into multiple sub-graphs, wherein each of (i) the at least one ontology graph and (ii) the at least one knowledge graph is based on both at least one user question and domain knowledge pertaining to the at least one user question, and wherein each state in the state automata represents a structured action comprising at least one of a query and a command;

generating a structured query for each of the multiple sub-graphs;

determining one or more intentions of a dialog pertaining to the at least one user question by translating each of the generated structured queries to a respective natural language query;

creating one or more dialog states for each of the determined dialog intentions;

creating one or more connecting dialog states between pairs of the created dialog states; and

generating an automata dialog framework associated with the at least one user question based on (i) the created dialog states and (ii) the created connecting dialog states.