Identifying common data objects representing solutions to a problem in different disciplines

• US 9,053,180 B2
• Filed: 06/11/2014
• Issued: 06/09/2015
• Est. Priority Date: 12/17/2009
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Rojahn, Tom O.
• Primary Examiner(s): Chaki, Kakali
• Assistant Examiner(s): Tran, Mai T
• Application Number:
  US14/301,413
• Publication Number:
  US 20140330871A1
• Time in Patent Office: 363 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06N 5/04 Inference or reasoning models

Abstract

An approach is provided for identifying a solution to a problem where the problem and solution are in different disciplines. First data elements in common between first and second ontologies are determined and second data elements in common between first and third ontologies are determined. Based on the number of first data elements exceeding a threshold value and the number of second data elements not exceeding the threshold value, a bridged ontology is generated that merges the first and second ontologies but not the first and third ontologies. A relationship is identified between a subject, predicate and object extracted from a first data object which specifies the problem and, respectively, a subject, predicate, and object of a second data object included in the bridged ontology. Based on the relationship, the solution is identified as being the second data object.

Claims

1. A computer system comprising:

a central processing unit (CPU);

a memory coupled to the CPU; and

a computer-readable storage device coupled to the CPU, the storage device containing instructions that are executed by the CPU via the memory to implement a method of identifying a solution to a problem, the problem and solution being in different first and second disciplines respectively represented by first and second ontologies, the method comprising the steps of;

in response to a determination that the first ontology represents the first discipline of the problem, the computer system retrieving the first ontology from a plurality of ontologies stored in a data store;

the computer system selecting the second ontology and a third ontology from the plurality of ontologies;

based on the second ontology being selected and in response to traversals of first and second directed graphs representing the first and second ontologies, respectively, the computer system determining a first number of first data elements in common between the first directed graph and the second directed graph, the first data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the second directed graph;

based on the third ontology being selected and in response to traversals of a third directed graph representing the third ontology and the first directed graph, the computer system determining a second number of second data elements in common between the first directed graph and the third directed graph, the second data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the third directed graph;

the computer system determining the first number of first data elements exceeds a threshold value and determining the second number of second data elements does not exceed the threshold value;

based on the first number of the first data elements in common between the first directed graph and the second directed graph exceeding the threshold value, and the second number of the second data elements in common between the first directed graph and the third directed graph not exceeding the threshold value, the computer system generating a bridged ontology that merges the first and second ontologies, but does not merge the first and third ontologies,

the computer system extracting a subject, predicate, and object of the problem from a first data object which specifies the problem, the first data object based on a determination of key words in a statement of the problem;

the computer system identifying a relationship between the extracted subject, predicate, and object and, respectively, a subject, predicate, and object of a second data object included in the bridged ontology; and

based on the identified relationship between the subject, predicate, and object extracted from the first data object which specifies the problem and, respectively, the subject, predicate, and object of the second data object included in the bridged ontology, the computer system identifying the solution to the problem as being the second data object.

2. The computer system of claim 1, wherein the method further comprises the steps of:

the computer system determining a first measure of dissimilarity between the first and second ontologies;

the computer system determining a second measure of dissimilarity between the first and third ontologies; and

the computer system determining the first and second measures of dissimilarity exceed a minimum measure of dissimilarity, wherein the step of selecting the second and third ontologies from the plurality of ontologies is based on the first and second measures of dissimilarity exceeding the minimum measure of dissimilarity.

3. The computer system of claim 2, wherein the step of generating the bridged ontology is based in part on the first measure of dissimilarity exceeding the minimum measure of dissimilarity.

4. The computer system of claim 1, wherein the method further comprises the steps of:

the computer system determining the key words in the statement of the problem;

based on the key words, the computer system generating the first data object specifying the problem; and

based on the first data object which specifies the problem, the computer system determining the first discipline of the problem.

5. The computer system of claim 1, wherein the method further comprises the step of the computer system determining the second data object includes the first data elements in common between the first and second directed graphs, wherein the step of identifying the solution to the problem as being the second data object is further based on the second data object including the first data elements in common between the first and second directed graphs.

6. The computer system of claim 1, wherein the method further comprises the steps of:

the computer system determining a plurality of data objects in the bridged ontology as potential solutions to the problem, the data objects consisting of respective subjects, predicates and objects;

the computer system identifying relationships between the subject, predicate, and object extracted from the first data object which specifies the problem and the respective subjects, predicates, and objects of which the data objects consist, wherein the step of identifying the relationships includes the step of identifying the relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object;

the computer system presenting a report including the identified relationships; and

in response to a review by a user of the relationships in the report, the computer system receiving from the user a confirmation that the identified relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object indicates the second data object is the solution to the problem.

7. The computer system of claim 1, wherein the step of identifying the relationship includes:

determining a word and a property of a data element including the subject of the first data object are identical to a word and a property of a data element including the subject of the second data object;

determining a word and a property of a data element including the predicate of the first data object are identical to a word and a property of a data element including the predicate of the second data object; and

determining a word and a property of a data element including the object of the first data object are identical to a word and a property of a data element including the object of the second data object.

8. A method for identifying a solution to a problem, the problem and solution being in different first and second disciplines respectively represented by first and second ontologies, the method comprising the steps of:

in response to a determination that the first ontology represents the first discipline of the problem, a computer retrieving the first ontology from a plurality of ontologies stored in a data store;

the computer selecting the second ontology and a third ontology from the plurality of ontologies;

based on the second ontology being selected and in response to traversals of the first and second directed graphs representing the first and second ontologies, respectively, the computer determining a first number of first data elements in common between the first directed graph and the second directed graph, the first data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the second directed graph;

based on the third ontology being selected and in response to a traversal of the first directed graph and a traversal of a third directed graph representing the third ontology, the computer determining a second number of second data elements in common between the first directed graph and the third directed graph, the second data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the third directed graph;

the computer determining the first number of first data elements exceeds a threshold value and determining the second number of second data elements does not exceed the threshold value;

based on the first number of the first data elements in common between the first directed graph and the second directed graph exceeding the threshold value, and the second number of the second data elements in common between the first directed graph and the third directed graph not exceeding the threshold value, the computer generating a bridged ontology that merges the first and second ontologies, but does not merge the first and third ontologies;

the computer extracting a subject, predicate, and object of the problem from a first data object which specifies the problem, the first data object based on a determination of key words in a statement of the problem;

the computer identifying a relationship between the extracted subject, predicate, and object and, respectively, a subject, predicate, and object of a second data object included in the bridged ontology; and

based on the identified relationship between the subject, predicate, and object extracted from the first data object which specifies the problem and, respectively, the subject, predicate, and object of the second data object included in the bridged ontology, the computer identifying the solution to the problem as being the second data object.

9. The method of claim 8, further comprising the steps of:

the computer determining a first measure of dissimilarity between the first and second ontologies;

the computer determining a second measure of dissimilarity between the first and third ontologies; and

the computer determining the first and second measures of dissimilarity exceed a minimum measure of dissimilarity, wherein the step of selecting the second and third ontologies from the plurality of ontologies is based on the first and second measures of dissimilarity exceeding the minimum measure of dissimilarity.

10. The method of claim 9, wherein the step of generating the bridged ontology is based in part on the first measure of dissimilarity exceeding the minimum measure of dissimilarity.

11. The method of claim 8, further comprising the steps of:

the computer determining the key words in the statement of the problem;

based on the key words, the computer generating the first data object specifying the problem; and

based on the first data object which specifies the problem, the computer determining the first discipline of the problem.

12. The method of claim 8, further comprising the step of the computer determining the second data object includes the first data elements in common between the first and second directed graphs, wherein the step of identifying the solution to the problem as being the second data object is further based on the second data object including the first data elements in common between the first and second directed graphs.

13. The method of claim 8, further comprising the steps of:

the computer determining a plurality of data objects in the bridged ontology as potential solutions to the problem, the data objects consisting of respective subjects, predicates and objects;

the computer identifying relationships between the subject, predicate, and object extracted from the first data object which specifies the problem and the respective subjects, predicates, and objects of which the data objects consist, wherein the step of identifying the relationships includes the step of identifying the relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object;

the computer presenting a report including the identified relationships; and

in response to a review by a user of the relationships in the report, the computer receiving from the user a confirmation that the identified relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object indicates the second data object is the solution to the problem.

14. The method of claim 8, wherein the step of identifying the relationship includes:

determining a word and a property of a data element including the subject of the first data object are identical to a word and a property of a data element including the subject of the second data object;

determining a word and a property of a data element including the predicate of the first data object are identical to a word and a property of a data element including the predicate of the second data object; and

determining a word and a property of a data element including the object of the first data object are identical to a word and a property of a data element including the object of the second data object.

15. A computer program product comprising:

a computer-readable storage medium; and

a computer-readable program code stored in the computer-readable storage medium, the computer-readable program code containing instructions that are executed by a central processing unit (CPU) of a computer system to implement a method of identifying a solution to a problem, the problem and solution being in different first and second disciplines respectively represented by first and second ontologies, the method comprising the steps of;

in response to a determination that the first ontology represents the first discipline of the problem, the computer system retrieving the first ontology from a plurality of ontologies stored in a data store;

the computer system selecting the second ontology and a third ontology from the plurality of ontologies;

based on the second ontology being selected and in response to traversals of first and second directed graphs representing the first and second ontologies, respectively, the computer system determining a first number of first data elements in common between the first directed graph and the second directed graph, the first data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the second directed graph;

based on the third ontology being selected and in response to traversals of a third directed graph representing the third ontology and the first directed graph, the computer system determining a second number of second data elements in common between the first directed graph and the third directed graph, the second data elements in common being data elements in the first directed graph whose words and properties are identical to respective words and properties in data elements in the third directed graph;

the computer system determining the first number of first data elements exceeds a threshold value and determining the second number of second data elements does not exceed the threshold value;

based on the first number of the first data elements in common between the first directed graph and the second directed graph exceeding the threshold value, and the second number of the second data elements in common between the first directed graph and the third directed graph not exceeding the threshold value, the computer system generating a bridged ontology that merges the first and second ontologies, but does not merge the first and third ontologies,

the computer system extracting a subject, predicate, and object of the problem from a first data object which specifies the problem, the first data object based on a determination of key words in a statement of the problem;

the computer system identifying a relationship between the extracted subject, predicate, and object and, respectively, a subject, predicate, and object of a second data object included in the bridged ontology; and

based on the identified relationship between the subject, predicate, and object extracted from the first data object which specifies the problem and, respectively, the subject, predicate, and object of the second data object included in the bridged ontology, the computer system identifying the solution to the problem as being the second data object.

16. The computer program product of claim 15, wherein the method further comprises the steps of:

the computer system determining a first measure of dissimilarity between the first and second ontologies;

the computer system determining a second measure of dissimilarity between the first and third ontologies; and

the computer system determining the first and second measures of dissimilarity exceed a minimum measure of dissimilarity, wherein the step of selecting the second and third ontologies from the plurality of ontologies is based on the first and second measures of dissimilarity exceeding the minimum measure of dissimilarity.

17. The computer program product of claim 16, wherein the step of generating the bridged ontology is based in part on the first measure of dissimilarity exceeding the minimum measure of dissimilarity.

18. The computer program product of claim 15, wherein the method further comprises the steps of:

the computer system determining the key words in the statement of the problem;

based on the key words, the computer system generating the first data object specifying the problem; and

based on the first data object which specifies the problem, the computer system determining the first discipline of the problem.

19. The computer program product of claim 15, wherein the method further comprises the step of the computer system determining the second data object includes the first data elements in common between the first and second directed graphs, wherein the step of identifying the solution to the problem as being the second data object is further based on the second data object including the first data elements in common between the first and second directed graphs.

20. The computer program product of claim 15, wherein the method further comprises the steps of:

the computer system determining a plurality of data objects in the bridged ontology as potential solutions to the problem, the data objects consisting of respective subjects, predicates and objects;

the computer system identifying relationships between the subject, predicate, and object extracted from the first data object which specifies the problem and the respective subjects, predicates, and objects of which the data objects consist, wherein the step of identifying the relationships includes the step of identifying the relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object;

the computer system presenting a report including the identified relationships; and

in response to a review by a user of the relationships in the report, the computer system receiving from the user a confirmation that the identified relationship between the extracted subject, predicate, and object and the subject, predicate, and object of the second data object indicates the second data object is the solution to the problem.