Providing answers to questions using hypothesis pruning

• US 9,323,831 B2
• Filed: 09/13/2012
• Issued: 04/26/2016
• Est. Priority Date: 09/28/2010
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chu-Carroll, Jennifer, Ferrucci, David A., Gondek, David C., Lally, Adam P., Murdock, James W. IV
• Primary Examiner(s): Trujillo, James
• Assistant Examiner(s): Le, Jessica N
• Application Number:
  US13/613,999
• Publication Number:
  US 20130018876A1
• Time in Patent Office: 1,321 Days
• Field of Search: 707/723
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24575 using context
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/3329 Natural language query formulation or dialogue systems
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation

Abstract

A method, system and computer program product for generating answers to questions. In one embodiment, the method comprises receiving a query, conducting a search through one or more data sources to identify candidate answers to the query, and providing each of the candidate answers with a preliminary score. The method further comprises filtering out any of the candidate answers with a preliminary score that does not satisfy a defined condition. The candidate answers having preliminary scores that satisfy this condition form a subset of the candidate answers. Each of the candidate answers in this subset is processed to produce further scores. A ranking function is applied to these further scores to determine a ranking for each of the candidate answers in the subset; and after this ranking function is applied, one or more of the candidate answers are selected as one or more final answers to the query.

Claims

1. A method of generating answers to questions, comprising:

receiving an input query;

conducting a search through one or more data sources to identify a plurality of candidate answers to the input query;

providing each of the candidate answers with a preliminary score based on one or more defined criterion;

filtering out any of the candidate answers with a preliminary score that does not satisfy a defined condition, wherein the candidate answers having preliminary scores that satisfy said defined condition form a subset of the candidate answers;

for each of the candidate answers in said subset, processing said each of the candidate answers to produce a plurality of further scores for said each of the candidate answers;

applying a candidate ranking function to said plurality of further scores to determine a ranking for said each of the candidate answers in said subset; and

after the applying the candidate ranking function, selecting one or more of the candidate answers as one or more final answers to the input query; and

wherein;

the processing said each of the candidate answers in the subset of candidate answers includes using a supporting passage retrieval process to identify the passages in which said each of the candidate answers occur;

the applying the candidate ranking function includes sending said subset of candidate answers from the supporting passage retrieval process to a candidate ranking function module, and using said candidate ranking function module to apply the candidate ranking function to said candidate answers in the subset of candidate answers; and

the processing said each of the candidate answers in the subset of candidate answers includes processing said each of the candidate answers using a context independent candidate answer process to obtain one or more post-filtering scores for said each of the candidate answers.

2. The method according to claim 1, wherein the applying a candidate ranking function includes ranking each of the candidate answers in the subset of candidate answers based on the post-filtering scores for said each of the candidate answers.

3. A method of generating answers to questions, comprising:

receiving an input query;

conducting a search through one or more data sources to identify a plurality of candidate answers to the input query, wherein each of the candidate answers is found in an associated supporting passage having identified content;

providing each of the candidate answers with a preliminary score based on one or more defined criterion;

filtering out any of the candidate answers with a preliminary score that does not satisfy a defined condition, wherein the candidate answers having preliminary scores that satisfy said defined condition form a subset of the candidate answers;

for each of the candidate answers in said subset, processing said each of the candidate answers to produce a plurality of further scores for said each of the candidate answers based on the content of the supporting passage in which said each candidate answers is found;

sending each of the candidate answers in said subset to a candidate ranking module;

using the candidate ranking module to apply a candidate ranking function to said plurality of further scores to determine a ranking for said each of the candidate answers in said subset;

sending said any of the filtered out candidate answers to the candidate ranking module;

using the candidate ranking module to determine a ranking for said any of the filtered out candidate answers; and

selecting one or more of the candidate answers, based on the ranking for the candidate answers, as one or more final answers to the input query; and

wherein;

the filtering out includes filtering out any of the candidate answers with a preliminary score that does not satisfy the defined condition prior to the processing said each of the candidate answers to produce a plurality of further scores for said each of the candidate answers based on the content of the supporting passage in which said each candidate answer is found;

the processing said each of the candidate answers in the subset of candidate answers includes using a supporting passage retrieval process to identify the passages in which said each of the candidate answers occur;

the sending each of the candidate answers in the subset of candidate answers to the ranking module includes sending said subset of candidate answers from the supporting passage retrieval process to the candidate ranking function module;

the processing said each of the candidate answers in the subset of candidate answers includes processing said each of the candidate answers using a context independent candidate answer process to obtain one or more post-filtering scores for said each of the candidate answers; and

the using the candidate ranking module to determine a ranking for said each of the candidate answers includes ranking said each of the candidate answers based on the post-filtering scores for said each of the candidate answers.