Answer interactions in a question-answering environment

• US 9,400,841 B2
• Filed: 05/11/2015
• Issued: 07/26/2016
• Est. Priority Date: 11/05/2014
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Eggebraaten, Thomas J., Megerian, Mark G., Petri, John E.
• Primary Examiner(s): Hill, Stanley K
• Assistant Examiner(s): OLUDE AFOLABI, OLATOYOSI
• Application Number:
  US14/708,689
• Publication Number:
  US 20160125075A1
• Time in Patent Office: 442 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24522 Translation of natural language queries to structured queries
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/3332 Query translation
• CPC Class Codes: G06F 16/3334 Selection or weighting of terms from queries, including natural language queries
• CPC Class Codes: G06F 16/3344 using natural language analysis
• CPC Class Codes: G06F 16/367 Ontology
• CPC Class Codes: G06F 16/90324 using system suggestions
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06F 16/9538 Presentation of query results
• CPC Class Codes: G06F 16/955 using information identifiers, e.g. uniform resource locators [URL]
• CPC Class Codes: G06F 40/205 Parsing
• CPC Class Codes: G06F 40/211 Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
• CPC Class Codes: G06F 40/284 Lexical analysis, e.g. tokenisation or collocates
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30
• CPC Class Codes: G06N 20/00 : Machine learning
• CPC Class Codes: G06N 5/022 : Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/046 : Forward inferencing; Production systems
• CPC Class Codes: G06N 7/01 : Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: G06Q 30/0203 : Market surveys; Market polls
• CPC Class Codes: G09B 7/00 : Electrically-operated teaching apparatus or devices working with questions and answers mechanically operated G09B3/00; computing arrangements G06F
• 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
• CPC Class Codes: G16H 10/20 : for electronic clinical trials or questionnaires
• CPC Class Codes: G16H 40/20 : for the management or administration of healthcare resources or facilities, e.g. managing hospital staff or surgery rooms
• CPC Class Codes: G16H 50/70 : for mining of medical data, e.g. analysing previous cases of other patients
• CPC Class Codes: G16Z 99/00 : Subject matter not provided for in other main groups of this subclass

Abstract

In a question-answering (QA) environment, a first answer sequence is identified. As identified, the first answer sequence includes a first answer and a second answer. A corpus is analyzed using the first answer and the second answer. Based on the analysis, a set of influence factors corresponding to both the first answer and the second answer are identified. A first answer relationship between the first answer and the second answer is then generated based on the set of influence factors.

Claims

1. A method for generating a first answer relationship in a first answer sequence of a plurality of answer sequences, the method comprising:

receiving, from a user, a request for instructions to achieve a result;

identifying, in response to the received request, the plurality of answer sequences, wherein each answer sequence of the plurality of answer sequences is a single, organized, complete answer that includes a plurality of answers to be used in combination to achieve the result;

identifying the first answer sequence of the plurality of answer sequences, the first answer sequence including a first answer and a second answer;

analyzing, using the first answer and the second answer, a corpus to identify a set of influence factors corresponding to both the first answer and the second answer, wherein each influence factor of the set of influence factors is a measure of compatibility between the first answer and the second answer when used in combination with each other; and

generating, based on the set of influence factors, the first answer relationship between the first answer and the second answer.

2. The method of claim 1, wherein the first answer sequence further includes a third answer, the method further comprising:

analyzing, using the third answer, the corpus to identify a second set of influence factors corresponding to both the first answer and the third answer, wherein each influence factor of the second set of influence factors is a measure of compatibility between the first answer and the third answer when used in combination with each other, and further to identify a third set of influence factors corresponding to both the second answer and the third answer, wherein each influence factor of the third set of influence factors is a measure of compatibility between the second answer and the third answer when used in combination with each other;

generating, based on the second set of influence factors, a second answer relationship between the first answer and the third answer; and

generating, based on the third set of influence factors, a third answer relationship between the first answer and the third answer.

3. The method of claim 2, further comprising:

evaluating, based on the first answer relationship, further based on the second answer relationship, and further based on the third answer relationship, the answer sequence.

4. The method of claim 1, further comprising:

assigning a relationship score to the first answer relationship, the relationship score calculated based on the set of influence factors; and

evaluating, based on the relationship score, the first answer relationship.

5. The method of claim 1, wherein the analyzing, using the first answer and the second answer, the corpus to identify the set of influence factors corresponding to both the first answer and the second answer includes:

identifying a first characteristic relationship between the first answer and a characteristic;

identifying a second characteristic relationship between the second answer and the characteristic; and

identifying, based on comparing the first characteristic relationship and the second characteristic relationship, a first influence factor of the set of influence factors.

6. The method of claim 1, wherein the analyzing, using the first answer and the second answer, the corpus to identify the set of influence factors corresponding to both the first answer and the second answer includes:

identifying a direct influence relationship between the first answer and the second answer; and

identifying, based on the direct influence relationship, a first influence factor of the set of influence factors.

7. The method of claim 1, wherein the analyzing, using the first answer and the second answer, the corpus to identify the set of influence factors corresponding to both the first answer and the second answer includes:

parsing, by a natural language processing technique configured to analyze syntactic and semantic content, the corpus.

8. The method of claim 1, further comprising:

assigning a first relationship score to the first answer relationship, the first relationship score calculated based on the set of influence factors;

assigning a first confidence score to the first answer sequence, the first confidence score based in part on the first relationship score; and

presenting, the first answer sequence to the user as a response to the received request.

20. The method of claim 1, wherein the set of influence factors include a one-direction influence factor and a two-direction influence factor, wherein the one-direction influence factor is an evaluation of an effect that the first answer is known to have on the second answer, and wherein the two-direction influence factor is an evaluation of an effect that the first answer and second answer are known to have on each other.

9. A method for generating a first answer relationship in a first answer sequence of a plurality of answer sequences, the method comprising:

receiving, from a user, a question;

identifying, in response to the received question, a plurality of answer sequences, wherein each answer sequence of the plurality of answer sequences is a procedure that includes as parts of that procedure a plurality of answers to be used together by the user to complete a task associated with the question;

identifying, in response to the received question, the first answer sequence of the plurality of answer sequences, the first answer sequence including a first answer and a second answer;

analyzing, using the first answer and the second answer, a corpus to identify a set of influence factors corresponding to both the first answer and the second answer, wherein each influence factor of the set of influence factors is an interaction that is likely to occur if the first answer and the second answer are used together as provided for in the first answer sequence to complete the task; and

generating, based on the set of influence factors, the first answer relationship between the first answer and the second answer.

10. The method of claim 9, wherein the first answer sequence further includes a third answer, the method further comprising:

analyzing, using the third answer, the corpus to identify a second set of influence factors corresponding to both the first answer and the third answer, wherein each influence factor of the second set of influence factors is an interaction that is likely to occur if the first answer and the third answer are used together as provided for in the first answer sequence to complete the task, and further to identify a third set of influence factors corresponding to both the second answer and the third answer, wherein each influence factor of the third set of influence factors is an interaction that is likely to occur if the second answer and the third answer are used together as provided for in the first answer sequence to complete the task;

generating, based on the second set of influence factors, a second answer relationship between the first answer and the third answer; and

generating, based on the third set of influence factors, a third answer relationship between the first answer and the third answer.

11. The method of claim 10, further comprising:

evaluating, based on the first answer relationship, further based on the second answer relationship, and further based on the third answer relationship, the answer sequence.

12. The method of claim 9, further comprising:

assigning a relationship score to the first answer relationship, the relationship score calculated based on the set of influence factors; and

evaluating, based on the relationship score, the first answer relationship.

13. The method of claim 9, wherein the analyzing, using the first answer and the second answer, the corpus to identify the set of influence factors corresponding to both the first answer and the second answer includes:

identifying a first characteristic relationship between the first answer and a characteristic;

identifying a second characteristic relationship between the second answer and the characteristic; and

identifying, based on comparing the first characteristic relationship and the second characteristic relationship, a first influence factor of the set of influence factors.

14. The method of claim 9, wherein the analyzing, using the first answer and the second answer, the corpus to identify the set of influence factors corresponding to both the first answer and the second answer includes:

identifying a direct influence relationship between the first answer and the second answer; and

identifying, based on the direct influence relationship, a first influence factor of the set of influence factors.

15. The method of claim 9, further comprising:

assigning a first relationship score to the first answer relationship, the first relationship score calculated based on the set of influence factors;

assigning a first confidence score to the first answer sequence, the first confidence score based in part on the first relationship score; and

presenting, the first answer sequence to the user as a response to the question.

16. The method of claim 9, wherein the set of influence factors include a one-direction influence factor and a two-direction influence factor, wherein the one-direction influence factor is an evaluation of an effect that the first answer is known to have on the second answer, and wherein the two-direction influence factor is an evaluation of an effect that the first answer and second answer are known to have on each other.

17. A method comprising:

receiving, from a user, an input query;

parsing, by a natural language processing technique configured to analyze syntactic and semantic content, the input query;

generating, based on the parsed input query and using a corpus, a plurality of answer sequences, wherein each answer sequence of the plurality of answer sequences is a plan that includes as parts of that plan a plurality of answers to be used together by the user;

identifying the first answer sequence from among the plurality of answer sequences, the first answer sequence including a first answer and a second answer;

analyzing, using the first answer and the second answer, the corpus to identify a set of influence factors corresponding to both the first answer and the second answer, wherein each influence factor of the set of influence factors is a measure of compatibility of the first answer and the second answer with each other; and

generating, based on the set of influence factors, the first answer relationship between the first answer and the second answer, wherein the first answer relationship is a composite of the set of influence factors and an indicator of how the first answer and the second answer are likely to influence the first answer sequence as a whole if the first answer sequence is used by the user;

assigning a relationship score to the first answer relationship, the relationship score calculated based on the set of influence factors;

determining that the relationship score is below a threshold; and

presenting, as a response to the input query, one or more of the plurality of answer sequences to the user, wherein the first answer sequence is omitted from the presented answer sequences based on the relationship score being below the threshold.

18. The method of claim 17, wherein the set of influence factors include a one-direction influence factor and a two-direction influence factor, wherein the one-direction influence factor is an evaluation of an effect that the first answer is known to have on the second answer, and wherein the two-direction influence factor is an evaluation of an effect that the first answer and second answer are known to have on each other.

19. The method 17, wherein the set of influence factors includes one influence factor that is inferred based on interactions between the first answer and a shared concept and between the second answer and the shared concept.