Correcting resource descriptive framework statements based on similarity of terms to web ontology

• US 10,013,499 B2
• Filed: 09/22/2014
• Issued: 07/03/2018
• Est. Priority Date: 09/22/2014
• Status: Active Grant

• Current Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Original Assignee: CA, Inc. (d/b/a CA Technologies) (Broadcom, Inc.)
• Inventors: Achyutuni, Kiran Jyotsna
• Primary Examiner(s): Lu, Kuen S
• Application Number:
  US14/492,472
• Publication Number:
  US 20160085877A1
• Time in Patent Office: 1,380 Days
• Field of Search: 707713, 715234
• CPC Class Codes: G06F 16/2453 Query optimisation
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06F 16/8358 Query translation
• CPC Class Codes: G06F 16/986 Document structures and storage, e.g. HTML extensions

Abstract

A list of unique terms, each having prefix and associated property data, contained in a defined web ontology is generated. A RDF statement about a web resource is received, and a list of unique terms contained in the RDF statement is generated. A problem term is identified within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in the web ontology. A list of candidate terms that are contained in the web ontology that satisfy a threshold similarity to the problem term is generated. A candidate term is selected from among the list of candidate terms having a data type for the property data matching a data type for the property data of the problem term. The candidate term is substituted for each occurrence of the problem term contained in the RDF statement.

Claims

1. A method by a computer comprising:

generating a list of unique terms, each comprising prefix data and associated property data, contained in a defined web ontology language;

receiving a resource descriptive framework (RDF) statement about a web resource;

generating a list of unique terms, each comprising prefix data and associated property data, contained in the RDF statement;

identifying a first problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in the defined web ontology language, the first problem term comprising an error in the RDF statement;

generating a list of candidate terms contained in the defined web ontology that satisfy a threshold similarity to the first problem term;

selecting a candidate term from among the list of candidate terms having a data type for the property data that matches a data type for the property data of the first problem term; and

substituting the candidate term for each occurrence of the first problem term contained in the RDF statement.

2. The method of claim 1, further comprising:

writing the RDF statement following the substituting to a web document in a memory that is readable by a networked computer.

3. The method of claim 2, wherein:

the receiving the RDF statement about the web resource, comprises reading the RDF statement from the web document in memory; and

the writing the RDF statement comprises replacing the RDF statement in the web document with the RDF statement following the substituting.

4. The method of claim 1, wherein the method is performed by a computer web page search engine, and further comprising:

writing the RDF statement following the substituting to a repository of semantic content of web pages from a plurality of websites.

5. The method of claim 1,

wherein receiving the RDF statement about the web resource, comprises:

receiving from a networked computer a request message containing a uniform resource identifier (URI) of a web document;

reading the RDF statement from the web document in memory; and

performing the substituting the candidate term for each occurrence of the first problem term contained in the RDF statement; and

further comprising communicating the RDF statement following the substituting to the network computer as part of a response RDF message to the request message.

6. The method of claim 1, further comprising:

prior to the selecting the candidate term from among the list of candidate terms, selecting a subset of the list of candidate terms that are within a list of historical candidate terms that have been earlier substituted for problem terms identified in other RDF statements,

wherein the selecting the candidate term from among the list of candidate terms having the data type for the property data that matches the data type for the property data of the first problem term, comprises selecting the candidate term from among the subset of the list of candidate terms.

7. The method of claim 6, further comprising:

identifying an author of the RDF statement; and

constraining members of the list of historical candidate terms to those that have been earlier substituted for problem terms identified in other RDF statements from the same author.

8. The method of claim 1, wherein the generating a list of candidate terms contained in the defined web ontology that satisfy the threshold similarity to the first problem term, comprises:

computing Levenshtein distance values between the first problem term and each of the list of candidate terms contained in the defined web ontology; and

including in the list of candidate terms the candidate terms having a computed Levenshtein distance value that satisfies a defined threshold value.

9. The method of claim 1, further comprising:

identifying a second problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in the defined web ontology, the second problem term comprising an error in the RDF statement;

determining that the defined web ontology contains a single candidate term that satisfies a threshold similarity to the second problem term; and

substituting the single candidate term for each occurrence of the second problem term contained in the RDF statement.

10. The method of claim 1, further comprising:

identifying a third problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in any of a plurality of defined web ontologies, the third problem term comprising an error in the RDF statement;

selecting one of the defined web ontologies based on identifying a match between the prefix data and the property data of the third problem term and prefix data and property data of a unique term contained in the selected one of the defined web ontologies;

substituting the unique term contained in the selected one of the defined web ontologies for each occurrence of the third problem term contained in the RDF statement.

11. A computer program product comprising:

a computer readable non-transitory storage medium having computer readable program code embodied in the non-transitory storage medium that when executed by a processor of a computer system causes the computer system to perform operations comprising;

generating a list of unique terms, each comprising prefix data and associated property data, contained in a defined web ontology language;

receiving a resource descriptive framework (RDF) statement about a web resource;

generating a list of unique terms, each comprising prefix data and associated property data, contained in the RDF statement;

identifying a first problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in the defined web ontology language, the first problem term comprising an error in the RDF statement;

generating a list of candidate terms contained in the defined web ontology that satisfy a threshold similarity to the first problem term;

selecting a candidate term from among the list of candidate terms having a data type for the property data that matches a data type for the property data of the first problem term; and

substituting the candidate term for each occurrence of the first problem term contained in the RDF statement.

12. The computer program product of claim 11, wherein the operations further comprise:

writing the RDF statement following the substituting to a web document in a memory that is readable by a networked computer.

13. The computer program product of claim 12, wherein:

the receiving the RDF statement about the web resource, comprises reading the RDF statement from the web document in memory; and

the writing the RDF statement comprises replacing the RDF statement in the web document with the RDF statement following the substituting.

14. The computer program product of claim 11, wherein the operations further comprise:

writing the RDF statement following the substituting to a repository of semantic content of web pages from a plurality of websites.

15. The computer program product of claim 11, wherein

receiving the RDF statement about the web resource, comprises:

receiving from a networked computer a request message containing a uniform resource identifier (URI) of a web document;

reading the RDF statement from the web document in memory; and

performing the substituting the candidate term for each occurrence of the first problem term contained in the RDF statement; and

the operations further comprise communicating the RDF statement following the substituting to the network computer as part of a response RDF message to the request message.

16. The computer program product of claim 11, wherein the operations further comprise:

prior to the selecting the candidate term from among the list of candidate terms, selecting a subset of the list of candidate terms that are within a list of historical candidate terms that have been earlier substituted for problem terms identified in other RDF statements,

wherein the selecting the candidate term from among the list of candidate terms having the data type for the property data that matches the data type for the property data of the first problem term, comprises selecting the candidate term from among the subset of the list of candidate terms.

17. The computer program product of claim 16, wherein the operations further comprise:

identifying an author of the RDF statement; and

constraining members of the list of historical candidate terms to those that have been earlier substituted for problem terms identified in other RDF statements from the same author.

18. The computer program product of claim 11, wherein the generating a list of candidate terms contained in the defined web ontology that satisfy the threshold similarity to the first problem term, comprises:

computing Levenshtein distance values between the first problem term and each of the list of candidate terms contained in the defined web ontology language; and

including in the list of candidate terms the candidate terms having a computed Levenshtein distance value that satisfies a defined threshold value.

19. The computer program product of claim 11, wherein the operations further comprise:

identifying a second problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in the defined web ontology language, the second problem term comprising an error in the RDF statement;

determining that the defined web ontology contains a single candidate term that satisfies a threshold similarity to the second problem term; and

substituting the single candidate term for each occurrence of the second problem term contained in the RDF statement.

20. The computer program product of claim 11, wherein the operations further comprise:

identifying a third problem term within the list of unique terms contained in the RDF statement that is not present among the list of unique terms contained in any of a plurality of defined web ontologies, the third problem term comprising an error in the RDF statement;

selecting one of the defined web ontologies based on identifying a match between the prefix data and the property data of the third problem term and prefix data and property data of a unique term contained in the selected one of the defined web ontologies;

substituting the unique term contained in the selected one of the defined web ontologies for each occurrence of the property data of the third problem term contained in the RDF statement.