Automatic data interpretation and answering analytical questions with tables and charts

Patent metadata

Field	Value
Publication number	US-9430558-B2
Application number	US-201514681594-A
Country	US
Kind code	B2
Filing date	Apr 8, 2015
Priority date	Sep 17, 2014
Publication date	Aug 30, 2016
Grant date	Aug 30, 2016

Abstract

A method providing an answer to at least one analytical question containing at least one table or at least one chart is provided. The method may include receiving an input question. The method may also include extracting a plurality of information from the input question based on a natural language analysis. The method may further include forming a well-defined sentence. The method may include extracting at least one table or at least one chart associated with the input question. The method may include forming at least one mathematical equation. The method may also include solving the at least one mathematical equation. The method may include determining the answer to the input question in natural language based on the solved at least one mathematical equation. The method may further include narrating the determined answer to the input question in natural language.

First claim

What is claimed is:

1. A method providing an answer to at least one analytical question containing at least one table or at least one chart, the method comprising:
receiving an input question containing at least one table or at least one chart, wherein the received input question is entered via a graphical user interface associated with a computer and the received input question is associated with the at least one analytical question;
extracting, by the computer, a plurality of information from the input question based on a natural language analysis of the input question;
forming a well-defined statement based on the extracted plurality of information;
extracting, by the computer, the at least one table or at least one chart associated with the input question;
extracting, by the computer, a plurality of information from the extracted at least one table or at least one chart, wherein the extracted plurality of information comprises a data value, a data label, a header, a footer, a legend, an overlay, a unit, a line, an edge, a contour, a shape, a length, a proportion, and an angle;
forming at least one mathematical equation based on the well-defined statement and the extracted plurality of information from the extracted at least one table or at least one chart;
solving the at least one mathematical equation;
determining the answer to the input question in natural language based on the solved at least one mathematical equation; and
narrating the determined answer to the input question in natural language, wherein narrating the determined answer to the input question comprises displaying the determined answer to the input question via the graphical user interface.

2. The method of claim 1 , wherein the extracted at least one table or at least one chart associated with the at least one analytical question is extracted in an image format.

3. The method of claim 2 , wherein the extracted image format includes extracted information associated with the extracted at least one table or at least one chart.

4. The method of claim 3 , wherein the extracted information comprises at least one of a data label, a header, a footer, a legend, an overlay, a data value or numeral, a unit, a line or an axis, a shape, a length, a proportion, and an angle.

5. The method of claim 3 , wherein the forming at least one mathematical equation comprises performing data interpretation on the extracted information.

6. The method of claim 2 , wherein the extracting the image format comprises the use of existing technology including at least one of pattern recognition technology and optical character recognition technology.

7. The method of claim 1 , wherein solving the at least one formed mathematical equation comprises using a standard mathematical formula.

8. The method of claim 1 , wherein the forming at least one mathematical equation and solving the at least one formed mathematical equation comprises using a dictionary containing a mapping between verbs and math operators.

Assignees

• IBM

Inventors

• Bhat Sandesh
• Mustafi Joy

Classifications

• G09B7/02 Primary

  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 title

• G06N5/022

  Knowledge engineering; Knowledge acquisition · CPC title

• G06T11/26

  Drawing of charts or graphs · CPC title

• G06F18/21

  Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation · CPC title

• G06F17/30654 Primary

  Physics · mapped topic