Browse Prior Art Database

Heuristics based context aware method to automatically convert large data table in several spreadsheets into representative visualizations/summary

IP.com Disclosure Number: IPCOM000219562D
Publication Date: 2012-Jul-06
Document File: 6 page(s) / 145K

Publishing Venue

The IP.com Prior Art Database

Abstract

The invention proposes a single-click print solution that conveniently allows users to print large data tables in a spreadsheet in very few pages, without losing any relevant context or information. This does so by extracting context of the data, establishing relationship among various data fields, and then automatically converting large table data into few relevant representative visualizations/ summary elements making printed output to consume very few pages and ink, which is also quick and convenient to understand. The proposed solution is divided into modules: M1 – Removing undesired columns of data M2 - Analyzing columns to find relations between columns a) Determining set of data and computed columns on the basis of functional relations. b) Determining contextual data labels from un-referred data columns using frequency distribution, data type and timestamps. c) Classifying columns into dimensions of visualization/ summary M3 - Determining visualization type as per the contextual information determined from previous steps.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 29% of the total text.

Page 01 of 6

Title

Heuristics based context aware method to automatically convert large data table in a spreadsheet into representative visualizations/summary.

Inventors/Authors

Naveen Goel Amit Mittal Himanshu Jindal

Summary

The invention proposes a heuristics based context aware solution to determine context of a data spreadsheet based on functional relations existing between data fields, data patterns and frequency distribution of data present in the spreadsheet while printing the sheet. Then information is represented in form of visualizations (like Charts/ Pivot Charts and Summary Tables) maintaining the context, allowing user to view, share and print the essence of entire data using minimal number of pages and ink.

Background

Spreadsheets with large data tables when printed consume lots of pages, which don't provide clear information because user has to dive deep into the table format to understand the data and for larger tables this problem becomes very time consuming and confusing.

Prior Art/Solutions

No system exists which makes claims as made by the invention.

There are solutions which use markup language to manipulate data's format and change chart views for specific vertical application segments. They are mainly template driven and work with a certain format of documents only within an assumed environment and context. Also there exist solutions which summarize/sum numbers or automatically create/draw charts. But none of these establishes relationship among various data column and leverage that context to dynamically generate representative visualizations/summary for the data table/set.

There exist Green Print products which make transformations to the document in a form optimized for printing. But they don't make any attempt to change the content (to extract essence of data), they just change the layout of the content.


Page 02 of 6

Description

The invention proposes a single-click print solution that conveniently allows users to print large data tables in a spreadsheet in very few pages, without losing any relevant context or information. This does so by extracting context of the data, establishing relationship among various data fields, and then automatically converting large table data into few relevant representative visualizations/ summary elements making printed output to consume very few pages and ink, which is also quick and convenient to understand.

Detailed Description:

The proposed solution is divided into modules:

M1 - Removing undesired columns of data

M2 - Analyzing columns to find relations between columns


a) Determining set of data and computed columns on the basis of functional relations.

b) Determining contextual data labels from un-referred data columns using frequency distribution, data type and timestamps.


c) Classifying columns into dimensions of visualization/ summary

M3 - Determining visualization type as per the contextual information determined from previous steps.

Below flowchart provides the order in whi...