Logic-based relationship graph expansion and extraction
Embodiments for relationship graph expansion and extraction from a collection of unstructured text data by a processor. A query relating to one or more concepts may be received. The query may be expanded according to a logical reasoning operation and a domain ontology having a set of logical rules. A relationship graph between one or more concepts from a plurality of unstructured text data may be extracted based on an expanded query according to a domain ontology and the set of logical rules.
1. A method, by a processor, for relationship graph expansion and extraction, comprising:
receiving, via a user interface of a computer, a user input query specifying a plurality of concepts;
searching a plurality of unstructured text data to generate relationships between the plurality of concepts, wherein generating the relationships includes expanding the user input query using a logical reasoning operation to infer the relationships between the one or more concepts notwithstanding whether the plurality of concepts are directly asserted to one another in the plurality of unstructured text data, and wherein inferring the relationships includes identifying, within the plurality of unstructured text data, a correlation between two or more indirect and independent statements being sub-components of a respective two explicit concepts of the plurality of concepts to generate a new concept not identified in the plurality of unstructured data prior to the searching and omitted from the user input query;
extracting a relationship graph between the plurality of concepts from the plurality of unstructured text data based on the expanded user input query according to a domain ontology, wherein the relationship graph is generated by:
recursively and incrementally determining, until a predefined number of paths has been reached, a shortest direct path between the plurality of concepts of the user input query,
for each of the predefined number of paths, extracting a logic formula from each respective path of the predefined number of paths by selecting a conjunction of types and relations represented by each edge of the plurality of concepts of the respective path, and
adding the logic formula representative of each respective path of the predefined number of paths to the domain ontology, wherein, the expanded user input query is re-executed against the plurality of unstructured text data subsequent to adding the logic formula; and
displaying one or more results comprising the expanded user input query and derived relationship graph on the user interface of the computer.
2. The method of claim 1 , further including expanding the user input query using the logical reasoning operation and the domain ontology having a set of logical rules.
3. The method of claim 1 , further including extracting the statements from the plurality of unstructured text data according to the expanded user input query with associated confidence scores.
4. The method of claim 1 , further including identifying a direct relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules.
5. The method of claim 1 , further including inferring an indirect relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules.
6. The method of claim 3 , further including asserting or inferring the statements with the associated confidence scores from the statements using the domain ontology and a set of logical rules.
7. The method of claim 1 , wherein extracting the relationship graph further includes combining both extracted and inferred relationships, and direct and indirect relationships between the plurality of concepts using the domain ontology and a set of logical rules.
8. A system for relationship graph expansion and extraction, comprising:
one or more computers with executable instructions that when executed cause the system to:
receive, via a user interface of the one or more computers, a user input query specifying a plurality of concepts;
search a plurality of unstructured text data to generate relationships between the plurality of concepts, wherein generating the relationships includes expanding the user input query using a logical reasoning operation to infer the relationships between the one or more concepts notwithstanding whether the plurality of concepts are directly asserted to one another in the plurality of unstructured text data, and wherein inferring the relationships includes identifying, within the plurality of unstructured text data, a correlation between two or more indirect and independent statements being sub-components of a respective two explicit concepts of the plurality of concepts to generate a new concept not identified in the plurality of unstructured data prior to the searching and omitted from the user input query;
extract a relationship graph between the plurality of concepts from the plurality of unstructured text data based on the expanded user input query according to a domain ontology, wherein the relationship graph is generated by:
recursively and incrementally determining, until a predefined number of paths has been reached, a shortest direct path between the plurality of concepts of the user input query,
for each of the predefined number of paths, extracting a logic formula from each respective path of the predefined number of paths by selecting a conjunction of types and relations represented by each edge of the plurality of concepts of the respective path, and
adding the logic formula representative of each respective path of the predefined number of paths to the domain ontology, wherein, the expanded user input query is re-executed against the plurality of unstructured text data subsequent to adding the logic formula; and
display one or more results comprising the expanded user input query and derived relationship graph on the user interface of the computer.
9. The system of claim 8 , wherein the executable instructions expand the user input query using the logical reasoning operation and the domain ontology having a set of logical rules.
10. The system of claim 8 , wherein the executable instructions extract the statements from the plurality of unstructured text data according to the expanded user input query with associated confidence scores.
11. The system of claim 8 , wherein the executable instructions identify a direct relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules.
12. The system of claim 8 , wherein the executable instructions infer an indirect relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules.
13. The system of claim 10 , wherein the executable instructions assert or infer the statements with the associated confidence scores from the statements using the domain ontology and a set of logical rules.
14. The system of claim 8 , wherein, pursuant to extracting the relationship graph, the executable instructions combine both extracted and inferred relationships, and direct and indirect relationships between the plurality of concepts using the domain ontology and a set of logical rules.
15. A computer program product for, by a processor, relationship graph expansion and extraction, the computer program product comprising a non-transitory computer-readable storage medium having computer-readable program code portions stored therein, the computer-readable program code portions comprising:
an executable portion that receives, via a user interface of a computer, a user input query specifying a plurality of concepts;
an executable portion that searches a plurality of unstructured text data to generate relationships between the plurality of concepts, wherein generating the relationships includes expanding the user input query using a logical reasoning operation to infer the relationships between the one or more concepts notwithstanding whether the plurality of concepts are directly asserted to one another in the plurality of unstructured text data, and wherein inferring the relationships includes identifying, within the plurality of unstructured text data, a correlation between two or more indirect and independent statements being sub-components of a respective two explicit concepts of the plurality of concepts to generate a new concept not identified in the plurality of unstructured data prior to the searching and omitted from the user input query;
an executable portion that extracts a relationship graph between the plurality of concepts from the plurality of unstructured text data based on the expanded user input query according to a domain ontology, wherein the relationship graph is generated by:
recursively and incrementally determining, until a predefined number of paths has been reached, a shortest direct path between the plurality of concepts of the user input query,
for each of the predefined number of paths, extracting a logic formula from each respective path of the predefined number of paths by selecting a conjunction of types and relations represented by each edge of the plurality of concepts of the respective path, and
adding the logic formula representative of each respective path of the predefined number of paths to the domain ontology, wherein, the expanded user input query is re-executed against the plurality of unstructured text data subsequent to adding the logic formula; and
an executable portion that displays one or more results comprising the expanded user input query and derived relationship graph on the user interface of the computer.
16. The computer program product of claim 15 , further including an executable portion that expands the user input query using the logical reasoning operation and the domain ontology having a set of logical rules.
17. The computer program product of claim 15 , further including an executable portion that extracts the statements from the plurality of unstructured text data according to the expanded user input query with associated confidence scores.
18. The computer program product of claim 15 , further including an executable portion that:
identifies a direct relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules; and
infers an indirect relationship from the statements between the plurality of concepts using the domain ontology and a set of logical rules.
19. The computer program product of claim 15 , further including an executable portion that asserts or infers the statements with the associated confidence scores from the statements using the domain ontology and a set of logical rules.
20. The computer program product of claim 17 , further including an executable portion that, pursuant to extracting the relationship graph, combines both extracted and inferred relationships, and direct and indirect relationships between the plurality of concepts using the domain ontology and a set of logical rules.