Question pruning for evaluating a hypothetical ontological link

• US 9,727,642 B2
• Filed: 11/21/2014
• Issued: 08/08/2017
• Est. Priority Date: 11/21/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Adderly, Darryl M., Allen, Corville O., Tucker, Robert K.
• Primary Examiner(s): Lee, Wilson
• Application Number:
  US14/549,602
• Publication Number:
  US 20160147875A1
• Time in Patent Office: 991 Days
• Field of Search: None
• CPC Class Codes: G06F 16/3346 using probabilistic model
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/90332 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

Mechanisms for generating a set of questions to evaluate a link between concept entities are provided. A hypothetical link between at least two information concept entities is generated. A set of questions corresponding to the hypothetical link is retrieved and pruned into a subset of questions based on at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities. The pruned set of questions is processed based on a corpus of evidence to generate a measure of support for or against the hypothetical link being an actual link between the at least two information concept entities. A validity indication for the hypothetical link indicating whether or not the hypothetical link is an actual link between the at least two information concept entities is output.

Claims

1. A method, in a data processing system, for generating a set of questions to evaluate a link between concept entities, comprising:

receiving, by the data processing system, a set of evidential data specifying a plurality of information concept entities;

analyzing, by the data processing system, an ontological data structure associated with the set of evidential data to identify potential ontological relationships between information concept entities in the plurality of information concept entities;

automatically generating, by the data processing system, a hypothetical link between at least two information concept entities in the plurality of information concept entities based on results of the analysis, wherein the hypothetical link represents a scenario involving the at least two information concept entities;

retrieving, by the data processing system, a set of questions corresponding to the hypothetical link between the at least two information concept entities;

pruning, by the data processing system, the set of questions into a subset of questions based on at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities;

processing, by the data processing system, the pruned set of questions based on a corpus of evidence to thereby generate a measure of support for or against the hypothetical link being an actual link between the at least two information concept entities; and

outputting, by the data processing system, an validity indication for the hypothetical link indicating whether or not the hypothetical link is an actual link between the at least two information concept entities.

2. The method of claim 1, wherein the set of questions are associated with a general scenario type associated with the hypothetical link, and wherein the pruned set of questions are specific to the particular scenario of the hypothetical link.

3. The method of claim 1, wherein the set of questions is generated by performing a lookup in a question template database of types of the information concept entities and a type of a relationship specified by the hypothetical link to identify a set of question templates corresponding to the types of information concept entities and type of relationship.

4. The method of claim 3, wherein pruning the set of questions into a subset of questions based on the at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities comprises:

comparing characteristics of the hypothetical link or characteristics of the at least two information concept entities to characteristics associated with question templates in the set of question templates; and

selecting question templates, from the set of question templates, that have at least one characteristic that matches at least one characteristic of the hypothetical link or at least one characteristic of the at least two information concept entities.

5. The method of claim 4, further comprising populating fields of the selected question templates with information from at least one of the information concept entities or the hypothetical link to generate the pruned set of questions.

6. The method of claim 4, wherein the characteristics of the at least two information concept entities comprise an occupation associated with each of the information concept entities, a classification of the information concept entity as being a victim, witness, bystander, or suspect, or a classification of a location represented by an information concept entity.

7. The method of claim 4, wherein the characteristics of the hypothetical link comprise a type of relationship between the information concept entities represented by the hypothetical link.

8. A computer program product comprising a computer readable storage medium having a computer readable program stored therein, wherein the computer readable program, when executed on a data processing system, causes the data processing system to:

receive a set of evidential data specifying a plurality of information concept entities;

analyze an ontological data structure associated with the set of evidential data to identify potential ontological relationships between information concept entities in the plurality of information concept entities;

automatically generate a hypothetical link between at least two information concept entities in the plurality of information concept entities based on results of the analysis, wherein the hypothetical link represents a scenario involving the at least two information concept entities;

retrieve a set of questions corresponding to the hypothetical link between the at least two information concept entities;

prune the set of questions into a subset of questions based on at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities;

process the pruned set of questions based on a corpus of evidence to thereby generate a measure of support for or against the hypothetical link being an actual link between the at least two information concept entities; and

output an validity indication for the hypothetical link indicating whether or not the hypothetical link is an actual link between the at least two information concept entities.

9. The computer program product of claim 8, wherein the set of questions are associated with a general scenario type associated with the hypothetical link, and wherein the pruned set of questions are specific to the particular scenario of the hypothetical link.

10. The computer program product of claim 8, wherein the set of questions is generated by performing a lookup in a question template database of types of the information concept entities and a type of a relationship specified by the hypothetical link to identify a set of question templates corresponding to the types of information concept entities and type of relationship.

11. The computer program product of claim 10, wherein the computer readable program causes the computing device to prune the set of questions into a subset of questions based on the at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities at least by:

comparing characteristics of the hypothetical link or characteristics of the at least two information concept entities to characteristics associated with question templates in the set of question templates; and

selecting question templates, from the set of question templates, that have at least one characteristic that matches at least one characteristic of the hypothetical link or at least one characteristic of the at least two information concept entities.

12. The computer program product of claim 11, wherein the computer readable program further causes the computing device to populate fields of the selected question templates with information from at least one of the information concept entities or the hypothetical link to generate the pruned set of questions.

13. The computer program product of claim 11, wherein the characteristics of the at least two information concept entities comprise an occupation associated with each of the information concept entities, a classification of the information concept entity as being a victim, witness, bystander, or suspect, or a classification of a location represented by an information concept entity.

14. The computer program product of claim 11, wherein the characteristics of the hypothetical link comprise a type of relationship between the information concept entities represented by the hypothetical link.

15. An apparatus comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to;

receive a set of evidential data specifying a plurality of information concept entities;

analyze an ontological data structure associated with the set of evidential data to identify potential ontological relationships between information concept entities in the plurality of information concept entities;

automatically generate a hypothetical link between at least two information concept entities in the plurality of information concept entities based on results of the analysis, wherein the hypothetical link represents a scenario involving the at least two information concept entities;

retrieve a set of questions corresponding to the hypothetical link between the at least two information concept entities;

prune the set of questions into a subset of questions based on at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities;

process the pruned set of questions based on a corpus of evidence to thereby generate a measure of support for or against the hypothetical link being an actual link between the at least two information concept entities; and

output an validity indication for the hypothetical link indicating whether or not the hypothetical link is an actual link between the at least two information concept entities.

16. The apparatus of claim 15, wherein the set of questions are associated with a general scenario type associated with the hypothetical link, and wherein the pruned set of questions are specific to the particular scenario of the hypothetical link.

17. The apparatus of claim 15, wherein the set of questions is generated by performing a lookup in a question template database of types of the information concept entities and a type of a relationship specified by the hypothetical link to identify a set of question templates corresponding to the types of information concept entities and type of relationship.

18. The apparatus of claim 17, wherein the instructions further cause the processor to prune the set of questions into a subset of questions based on the at least one of characteristics of the hypothetical link or characteristics of the at least two information concept entities at least by:

comparing characteristics of the hypothetical link or characteristics of the at least two information concept entities to characteristics associated with question templates in the set of question templates; and

selecting question templates, from the set of question templates, that have at least one characteristic that matches at least one characteristic of the hypothetical link or at least one characteristic of the at least two information concept entities.

19. The apparatus of claim 18, further comprising populating fields of the selected question templates with information from at least one of the information concept entities or the hypothetical link to generate the pruned set of questions.

20. The apparatus of claim 18, wherein the characteristics of the at least two information concept entities comprise at least one of an occupation associated with each of the information concept entities, a classification of the information concept entity as being a victim, witness, bystander, or suspect, a classification of a location represented by an information concept entity, or a type of relationship between the information concept entities represented by the hypothetical link.