Content revision using question and answer generation

• US 9,934,220 B2
• Filed: 04/10/2014
• Issued: 04/03/2018
• Est. Priority Date: 08/09/2012
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Jenkins, Jana H, Steinmetz, David C, Zadrozny, Wlodek W
• Primary Examiner(s): Alam, Hosain
• Assistant Examiner(s): ALLEN, NICHOLAS E
• Application Number:
  US14/250,128
• Publication Number:
  US 20140222822A1
• Time in Patent Office: 1,454 Days
• CPC Class Codes: G06F 16/38 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
• CPC Class Codes: G06F 16/93 Document management systems
• CPC Class Codes: G06F 40/56 Natural language generation

Abstract

A method for question/answer creation for a document is described. The method includes importing a document having a set of questions based on content in the document. The method also includes automatically creating a candidate question from the content in the document. The method also includes automatically generating answers for the set of questions and the candidate question using the content in the document. The method also includes presenting the set of questions, the candidate question, and the answers to a content creator for user verification of accuracy. The method also includes storing a verified set of questions in the document. The verified set of questions includes the candidate question.

Claims

1. A method for question/answer creation for a document, comprising:

importing, using a processor, a document created by a content creator, the document having a set of questions based on content in the document, wherein the document is a single file;

scanning, using a processor, the content of the document and metadata in the document;

automatically creating, using a processor, a candidate question not entered by the content creator, creating the candidate question from the content in the document and the metadata of the document, wherein the metadata is not visible when the document is opened by a content user;

automatically generating, using a processor, answers for the set of questions and the candidate question using the content in the document;

presenting, using a processor, the set of questions, the candidate question, and the answers to a content creator for user verification of accuracy; and

storing, using a processor, a verified set of questions in the document, wherein the verified set of questions comprises at least one verified question from the set of questions and the candidate question.

2. The method of claim 1, wherein automatically creating the candidate question further comprises:

detecting, using a processor, a pattern in the content of the document that is common for other content in a corpus to which the document belongs; and

creating, using a processor, the candidate question based on the pattern.

3. The method of claim 1, wherein importing the document further comprises:

scanning, using a processor, viewable content of the document and metadata in the document to obtain the set of questions; and

categorizing, using a processor, the set of questions based on the content of the document.

4. The method of claim 1, further comprising:

verifying, using a processor, the set of questions, the candidate question, and the answers for accuracy according to the content of the document and applicability to the content of the document; and

revising, using a processor, the candidate question to correctly reflect the content in the document.

5. The method of claim 1, further comprising:

scoring, using a processor, question/answer pairs generated by the system, wherein question/answer pairs that meet a scoring threshold are retained.

6. The method of claim 1, wherein storing the verified set of questions in the document further comprises:

populating, using a processor, the verified set of questions with the candidate question in response to determining that the candidate question is accurate based on the content.

7. The method of claim 1, further comprising:

receiving, using a processor, feedback from a user, wherein the feedback is related to the document;

receiving, using a processor, an input from the content creator to create a new question corresponding to the content in the document and based on the feedback; and

automatically generating, using a processor, answers for the new question using the content in the document.