System and method for computerized semantic indexing and searching

• US 10,678,820 B2
• Filed: 04/11/2019
• Issued: 06/09/2020
• Est. Priority Date: 04/12/2018
• Status: Active Grant

• Current Assignee: Abel Browarnik
• Original Assignee: Abel Browarnik
• Inventors: Browarnik, Abel
• Primary Examiner(s): Yen, Eric
• Application Number:
  US16/381,024
• Publication Number:
  US 20190317953A1
• Time in Patent Office: 425 Days
• CPC Class Codes: G06F 16/316 Indexing structures
• CPC Class Codes: G06F 16/328 Management therefor
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/247 Thesauruses; Synonyms
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/289 Phrasal analysis, e.g. finite state techniques or chunking
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/35 Discourse or dialogue representation

Abstract

A semantic indexing system, the semantic indexing system comprising a processing resource configured to: provide a corpus comprising a plurality of textual documents, wherein (a) each of the textual documents being composed of one or more sentences; (b) each of the sentences being composed of one or more statements; and generate an index, the index mapping each of the statements to one or more frames; wherein each frame defines a structure that carries a semantic meaning, thereby enabling searching the corpus by the semantic meaning of a search statement.

Claims

1. A semantic indexing system, the semantic indexing system comprising a processing device configured to:

provide a corpus comprising a plurality of textual documents, wherein (a) each of the textual documents being composed of one or more sentences;

(b) each of the sentences being composed of one or more statements; and

generate an index, the index mapping each of the statements to one or more corresponding frames of a plurality of unique frames;

wherein;

(a) each frame of the plurality of unique frames is a pattern consisting of a syntactic description and one or more semantic predicates with a temporal function, and mapping each of the statements to the corresponding frames is performed using a verb of the corresponding statement and using VerbNet, or (b) each frame of the plurality of unique frames represents a situation involving one or more participants, one or more props, and one or more conceptual roles, and mapping each of the statements to the corresponding frames is performed using the verb of the corresponding statement and using FrameNet; and

wherein each frame of the plurality of unique frames defines a structure that carries a semantic meaning, thereby enabling searching the corpus by a semantic meaning of a search statement.

2. The semantic indexing system of claim 1, wherein the processing device is further configured to:

receive a given search statement;

map at least one of;

(a) the given search statement, and (b) one or more derivatives thereof, to one or more of the plurality of unique frames, giving rise to search frames; and

generate, using the index, a list of one or more of the textual documents comprising one or more selected statements of the statements, wherein the selected statements are mapped to at least one of the search frames.

3. The semantic indexing system of claim 2, wherein the given search statement is provided by a user and wherein the processing device is further configured to:

provide the list to the user;

receive, from the user, information enabling identification of suspected incompatibilities between (a) suspected incompatible statements of the selected statements within the textual documents within the list and (b) the given search statement; and

perform one or more suspected incompatibilities resolution measures to resolve the suspected incompatibilities.

4. The semantic indexing system of claim 3, wherein a given suspected incompatibilities resolution measure of the suspected incompatibilities resolution measures includes:

providing a second user with the information enabling identification of the suspected incompatibilities;

receiving, from the second user, a suspected incompatibilities resolution instruction for updating the index; and

updating the index in accordance with the suspected incompatibilities resolution instruction received from the second user.

5. The semantic indexing system of claim 3, wherein a given suspected incompatibilities resolution measure of the suspected incompatibilities resolution measures includes automatically removing the mapping between the suspected incompatible statements and the search frames.

6. The semantic indexing system of claim 2, wherein the one or more derivatives are determined by replacing one or more words of the given search statement with semantically-related replacement words.

7. The semantic indexing system of claim 1, wherein the processing device is further configured to perform the following for each of the statements that are mappable to a plurality of given frames identified by a given mapping scheme:

calculate a compatibility score indicative of a compatibility of each of the given frames to the respective statement; and

select one or more of the given frames, utilizing the compatibility score, giving rise to mapped given frames, being the given frames that the respective statement is mapped to.

8. The semantic indexing system of claim 7, wherein the compatibility score is determined by a correlation between statement elements of the respective statement and frame elements of the respective given frame.

9. The semantic indexing system of claim 8, wherein the statement elements and the frame elements are associated with respective element types and wherein the correlation is weighted in accordance with a weight given to each of the element types.

10. The semantic indexing system of claim 7, wherein the given mapping scheme is based on at least one of (a) VerbNet and (b) FrameNet.

11. A semantic indexing method, the semantic indexing method comprising:

providing, by a processing resource, a corpus comprising a plurality of textual documents, wherein (a) each of the textual documents being composed of one or more sentences;

(b) each of the sentences being composed of one or more statements; and

generating, by the processing resource, an index, the index mapping each of the statements to one or more corresponding frames of a plurality of unique frames;

wherein;

(a) each frame of the plurality of unique frames is a pattern consisting of a syntactic description and one or more semantic predicates with a temporal function, and mapping each of the statements to the corresponding frames is performed using a verb of the corresponding statement and using VerbNet, or (b) each frame of the plurality of unique frames represents a situation involving one or more participants, one or more props, and one or more conceptual roles, and mapping each of the statements to the corresponding frames is performed using the verb of the corresponding statement and using FrameNet; and

wherein each frame of the plurality of unique frames defines a structure that carries a semantic meaning, thereby enabling searching the corpus by a semantic meaning of a search statement.

12. The semantic indexing method of claim 11, further comprising:

receiving, by the processing resource, a given search statement;

mapping, by the processing resource, at least one of;

(a) the given search statement, and (b) one or more derivatives thereof, to one or more of the plurality of unique frames, giving rise to search frames; and

generating, by the processing resource, using the index, a list of one or more of the textual documents comprising one or more selected statements of the statements, wherein the selected statements are mapped to at least one of the search frames.

13. The semantic indexing method of claim 12, wherein the given search statement is provided by a user and wherein the method further comprises:

providing the list to the user;

receiving, by the processing resource, from the user, information enabling identification of suspected incompatibilities between (a) suspected incompatible statements of the selected statements within the textual documents within the list and (b) the given search statement; and

performing, by the processing resource, one or more suspected incompatibilities resolution measures to resolve the suspected incompatibilities.

14. The semantic indexing method of claim 13, wherein a given suspected incompatibilities resolution measure of the suspected incompatibilities resolution measures includes:

providing a second user with the information enabling identification of the suspected incompatibilities;

receiving, from the second user, a suspected incompatibilities resolution instruction for updating the index; and

updating the index in accordance with the suspected incompatibilities resolution instruction received from the second user.

15. The semantic indexing method of claim 13, wherein a given suspected incompatibilities resolution measure of the suspected incompatibilities resolution measures includes automatically removing the mapping between the suspected incompatible statements and the search frames.

16. The semantic indexing method of claim 12, wherein the one or more derivatives are determined by replacing one or more words of the given search statement with semantically-related replacement words.

17. The semantic indexing method of claim 11, further comprising performing, by the processing resource, the following for each of the statements that are mappable to a plurality of given frames identified by a given mapping scheme:

calculating a compatibility score indicative of a compatibility of each of the given frames to the respective statement; and

selecting one or more of the given frames, utilizing the compatibility score, giving rise to mapped given frames, being the given frames that the respective statement is mapped to.

18. The semantic indexing method of claim 17, wherein the compatibility score is determined by a correlation between statement elements of the respective statement and frame elements of the respective given frame.

19. The semantic indexing method of claim 17, wherein the given mapping scheme is based on at least one of (a) VerbNet and (b) FrameNet.

20. A non-transitory computer readable storage medium having computer readable program code embodied therewith, the computer readable program code, executable by at least one processor of a computer to perform a method comprising:

providing, by the at least one processor, a corpus comprising a plurality of textual documents, wherein (a) each of the textual documents being composed of one or more sentences;

(b) each of the sentences being composed of one or more statements; and

generating, by the at least one processor, an index, the index mapping each of the statements to one or more corresponding frames of a plurality of unique frames;

wherein;

(a) each frame of the plurality of unique frames is a pattern consisting of a syntactic description and one or more semantic predicates with a temporal function, and mapping each of the statements to the corresponding frames is performed using a verb of the corresponding statement and using VerbNet, or (b) each frame of the plurality of unique frames represents a situation involving one or more participants, one or more props, and one or more conceptual roles, and mapping each of the statements to the corresponding frames is performed using the verb of the corresponding statement and using FrameNet; and

wherein each frame of the plurality of unique frames defines a structure that carries a semantic meaning, thereby enabling searching the corpus by a semantic meaning of a search statement.