Document preparation with argumentation support from a deep question answering system

• US 10,387,576 B2
• Filed: 11/30/2017
• Issued: 08/20/2019
• Est. Priority Date: 11/30/2017
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Beller, Charles E., Dubyak, William G., Sakthi, Palani, Sheets, Robert A., Summers, Kristen M.
• Primary Examiner(s): Yen, Eric
• Application Number:
  US15/827,016
• Publication Number:
  US 20190163745A1
• Time in Patent Office: 628 Days
• 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
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06F 40/51 Translation evaluation
• CPC Class Codes: G09B 19/00 Teaching not covered by other main groups of this subclass teaching or practice apparatus for gun-aiming or gun-laying F41G3/26
• CPC Class Codes: G09B 5/02 with visual presentation of the material to be studied, e.g. using film strip
• CPC Class Codes: G09B 7/02 of the type wherein the student is expected to construct an answer to the question which is presented or wherein the machine gives an answer to the question presented by a student

Abstract

Dynamic semantic processing of text in a word processing application with engagement of question-answering system. A user provides a text stream to a computer system via an input source. The input text stream includes a first natural language statement. The system determines that the first natural language statement includes a fact-based component expressed in natural language form. The system identifies an initial span and an alternative span of the first natural language component, based on the determining. The system engages a question-answering (QA) system by providing the initial span and the alternative span of the first natural language component to the QA system, and by receiving, in response to the providing, a set of natural language results from the QA system. The system evaluates the initial span and the alternative span of the first natural language component based on the received natural language results.

Claims

1. A method for processing user input in a word processing program, comprising:

receiving an input text stream from a user via an input source, the input text stream comprising a first natural language statement;

determining that the first natural language statement comprises a fact-based first natural language component expressed in natural language form;

identifying an initial span and an alternative span of the first natural language component, based on the determining;

engaging a question-answering (QA) system by providing the initial span and the alternative span of the first natural language component to the QA system, and by receiving, in response to the providing, a set of natural language results from the QA system; and

evaluating the initial span and the alternative span of the first natural language component based on the received natural language results.

2. The method of claim 1, wherein the natural language results comprise a passage, a keyword, or both.

3. The method of claim 2, wherein each of the passage, the keyword, or both, is associated with a score reflecting how well the passage, the keyword, or both, supports an associated span of text, the associated span of text corresponding to the initial span or the alternative span of the first natural language component.

4. The method of claim 2, wherein engaging a QA system further comprises receiving a set of scores associated with the passage, a plurality of keyword results, or both.

5. The method of claim 1, further comprising:

identifying, based on the evaluating, a set of passages and keywords having associated scores corresponding to how well the set of passages and keywords support the initial span and the alternative span of the first natural language component, wherein the associated scores are within a specified range of scores for passages and keywords.

6. The method of claim 1, further comprising:

outputting, to a user, results of the evaluating via an input/output (I/O) device of a computing device.

7. The method of claim 6, wherein outputting results of the evaluating via an input/output (I/O) device of the computing device comprises:

highlighting, on a display of the computing device, the initial span of the first natural language component; and

indicating that at least one alternative text span of the first natural language component is associated with a set of passages, keywords, or both, having corresponding scores higher than scores for the initial span of the first natural language component.

8. The method of claim 1, wherein receiving an input text stream via an input source comprises:

monitoring an input stream of the word processing program based on keyboard input from a user; and

wherein, determining that the first natural language statement comprises a fact-based component comprises

periodically engaging the QA system using, as input to the QA system, a portion of the input stream of the word processing program.

9. The method of claim 1, wherein evaluating the initial span and the alternative span of the first natural language component based on the received natural language results comprises:

determining which of the initial span and the alternative span of the first natural language component has a higher associated score returned by the QA system.

10. A computer program product for processing user input in a word processing program, the computer program product comprising a non-transitory tangible storage device having program code embodied therewith, the program code executable by a processor of a computer to perform a method, the method comprising:

receiving, by the processor, an input text stream from a user via an input source, the input text stream comprising a first natural language statement;

determining, by the processor, that the first natural language statement comprises a fact-based first natural language component expressed in natural language form;

identifying, by the processor, an initial span and an alternative span of the first natural language component, based on the determining;

engaging, by the processor, a question-answering (QA) system by providing the initial span and the alternative span of the first natural language component to the QA system, and by receiving, in response to the providing, a set of natural language results from the QA system; and

evaluating, by the processor, the initial span and the alternative span of the first natural language component based on the received natural language results.

11. The computer program product of claim 10, wherein the natural language results comprise a passage, a keyword, or both.

12. The computer program product of claim 11, wherein each of the passage, the keyword, or both, is associated with a score reflecting how well the passage, the keyword, or both, supports an associated span of text, the associated span of text corresponding to the initial span or the alternative span of the first natural language component.

13. The computer program product of claim 11, wherein engaging a QA system further comprises receiving a set of scores associated with the passage, a plurality of keyword results, or both.

14. The computer program product of claim 10, further comprising:

identifying, by the processor, based on the evaluating, a set of passages and keywords having associated scores corresponding to how well the set of passages and keywords support the initial span and the alternative span of the first natural language component, wherein the associated scores are within a specified range of scores for passages and keywords.

15. The computer program product of claim 10, further comprising:

outputting, by the processor, to a user, results of the evaluating via an input/output (I/O) device of a computing device.

16. The computer program product of claim 15, wherein outputting results of the evaluating via an input/output (I/O) device of the computing device comprises:

highlighting, by the processor, on a display of the computing device, the initial span of the first natural language component; and

indicating, by the processor, that at least one alternative text span of the first natural language component is associated with a set of passages, keywords, or both, having corresponding scores higher than scores for the initial span of the first natural language component.

17. The computer program product of claim 10, wherein receiving an input text stream via an input source comprises:

monitoring an input stream of the word processing program based on keyboard input from a user; and

wherein, determining that the first natural language statement comprises a fact-based component comprises

periodically engaging the QA system using, as input to the QA system, a portion of the input stream of the word processing program.

18. A computer system for processing user input in a word processing program, comprising:

one or more computer devices each having one or more processors and one or more tangible storage devices; and

a program embodied on at least one of the one or more storage devices, the program having a plurality of program instructions for execution by the one or more processors, the program instructions comprising instructions for

receiving an input text stream from a user via an input source, the input text stream comprising a first natural language statement;

determining that the first natural language statement comprises a fact-based first natural language component expressed in natural language form;

identifying an initial span and an alternative span of the first natural language component, based on the determining;

engaging a question-answering (QA) system by providing the initial span and the alternative span of the first natural language component to the QA system, and by receiving, in response to the providing, a set of natural language results from the QA system;

evaluating the initial span and the alternative span of the first natural language component based on the received natural language results; and

identifying, based on the evaluating, a set of passages and keywords having associated scores corresponding to how well the set of passages and keywords support the initial span and the alternative span of the first natural language component, wherein the associated scores are within a specified range of scores for passages and keywords.

19. The system of claim 18, wherein the natural language results comprise a passage, a keyword, or both.

20. The system of claim 19, wherein each of the passage, the keyword, or both, is associated with a score reflecting how well the passage, the keyword, or both, supports an associated span of text, the associated span of text corresponding to the initial span or the alternative span of the first natural language component.