Adapting tabular data for narration

• US 10,303,741 B2
• Filed: 11/26/2013
• Issued: 05/28/2019
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Byron, Donna K., Pikovsky, Alexander, Sanchez, Matthew B.
• Primary Examiner(s): Baderman, Scott T
• Assistant Examiner(s): Golden, Steve P
• Application Number:
  US14/089,953
• Publication Number:
  US 20140281935A1
• Time in Patent Office: 2,009 Days
• CPC Class Codes: G06F 40/103 Formatting, i.e. changing of presentation of documents automatic justification G06F40/189; automatic line break hyphenation G06F40/191
• CPC Class Codes: G06F 40/177 of tables; using ruled lines
• CPC Class Codes: G06F 40/56 Natural language generation

Abstract

A method for adapting tabular data for narration is provided in the illustrative embodiments. A set of categories used to organize data is identified in a first tabular portion of a document. A structure of the categories is analyzed. An inference is drawn about data in a first cell in the first tabular portion based on a position of the first cell in the structure. The first tabular portion of the document is transformed into a first narrative form using the inference.

Claims

1. A method for adapting tabular data for narration, the method comprising:

identifying, in a first tabular portion of a document, a set of categories used to organize data;

analyzing a structure of the categories, wherein the set of categories is arranged in a hierarchy in the first tabular portion;

drawing an inference of inclusion between a plurality of clauses corresponding to a plurality of categories in the hierarchy, the inference determining that one clause in the plurality of clauses includes another clause in the plurality of clauses according to a natural language-dependent grammatical rule;

connecting the plurality of clauses in the hierarchy into a narrative statement according to linguistic dependencies between the plurality of clauses, the plurality of clauses corresponding to the plurality of categories from the set of categories;

drawing an inference about data in a first cell in the first tabular portion based on a position of the first cell in the structure; and

transforming the first tabular portion of the document into a first narrative form using the inference, wherein the first narrative form uses the narrative statement.

2. The method of claim 1, wherein the transforming comprises:

creating an inferential narrative, wherein the inferential narrative comprises inferences drawn using cells in different tabular portions of the document, the inferential narrative describing content not present in the different tabular portions.

3. The method of claim 1, wherein the drawing the inference further comprises:

cross-referencing the data in the first cell with data in a non-tabular portion of the document.

4. The method of claim 1, further comprising:

merging the first narrative form corresponding to the first tabular portion with a non-tabular portion of the document.

5. The method of claim 1, further comprising:

sending the narrative form of the first tabular portion of the document and a non-tabular portion of the document to a natural language processing engine.

6. The method of claim 1, further comprising:

receiving the document, the document including a plurality of tabular portions, the first tabular portion being one of the plurality of tabular portions, each tabular portion organizing data using a different set of categories.

7. The method of claim 1, wherein the transforming generates the first narrative form without using a template for a narrative sentence, and wherein the transforming comprises:

determining a significance of the first cell relative to the first tabular data based on the position of the first cell in the structure; and

generating a part of the first narrative form using the determined significance of the first cell.

8. The method of claim 7, wherein the first cell does not include a value, wherein the first cell specifies a category, and wherein the first narrative form describes the category in the first cell.

9. The method of claim 1, wherein the drawing the inference further comprises:

cross-referencing the data in the first cell with data in a second cell in the first tabular portion.

10. The method of claim 9, wherein the cross-referencing further comprises:

using cross-referencing information created during transforming a second document.