Browse Prior Art Database

Question Chaining : Method for asking follow-up questions to a previous question in a similar context to refine a set of answers from the same answer bucket in a natural language question answer (NLQA) system.

IP.com Disclosure Number: IPCOM000237479D
Publication Date: 2014-Jun-18
Document File: 2 page(s) / 81K

Publishing Venue

The IP.com Prior Art Database

Abstract

A method for asking follow-up questions to a previous question in a similar context to refine a set of answers from the same answer bucket in a natural language question answer (NLQA) system is disclosed.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 51% of the total text.

Page 01 of 2

Question Chaining : Method for asking follow-up questions to a previous question in a similar context to refine a set of answers from the same answer bucket in a natural language question answer (NLQA) system.

Disclosed is a method for asking follow-up questions to a previous question in a similar context to refine a set of answers from the same answer bucket in a natural language question answer (NLQA) system.

Current search engines typically generate new sets of answers each time a question is asked even though the context for the next question has not changed. This occurs when a follow-up question is entered from the previous question. This approach presents more than the required information. Methods to narrow information are usually based on filters that are not fine-grained enough. In many cases, they may eliminate too much information from the answers. Hence, users of search engines asking related questions are unable to build on what they learned from previous questions. They are unable to keep an answer bucket constant so that follow-up questions can be used as a source for answers from the same bucket.

The disclosed method utilizes a question chain that allows users who are asking follow-up questions to refine the answers. The method supports users asking questions in a similar context to be able to refine their answers based on the answers that were found in the previous question. The method allows users to interact with the question chain in a simple way that gives them control over what results they are gett...