Automated Master Truth Data Construction for Production Data Evaluation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Evaluating the accuracy of large quantities of production data is prohibitively expensive due to the high time and effort required for human data entry, leading to compromises in data evaluation quality and quantity.

Innovation Solution

The construction of master truth data sets by comparing production data sets and provisional truth data sets, which are treated as statistically independent, using different protocols for interpretation, allowing for automated processing and reduced costs while maintaining high accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If human data entry personnel manually verify production data to determine truth, then measurement precision is improved, but productivity deteriorates due to prohibitively high time and effort requirements

Engineering Contradiction:

Improveaccuracy of production data evaluationVSAvoidspeed of constructing master truth data sets

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces the mechanical system of manual human data entry and verification with an automated computer-based system that uses optical character recognition (OCR) and image processing technologies to extract and verify data from source documents, thereby maintaining measurement precision while dramatically improving productivity

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent creates digital copies (images) of source documents and processes these copies through automated recognition systems to extract data, eliminating the need for manual copying and data entry while preserving the accuracy of the original documents

Inventive Principle:

Principle #26Copying

2Measurement precision

If human personnel double-key verification is performed to achieve desired statistical accuracy, then measurement precision is improved, but loss of time increases to prohibitively expensive levels

Engineering Contradiction:

Improvestatistical accuracy of production dataVSAvoidtime required for evaluating production data

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary automated data extraction and verification using OCR and image processing before final evaluation, preparing the data in advance with high accuracy so that subsequent statistical analysis can be performed quickly without requiring extensive manual verification time

Inventive Principle:

Principle #10Preliminary action

3Measurement precision

If the amount of production data evaluated is increased to improve statistical accuracy, then measurement precision is improved, but productivity deteriorates due to the linear increase in human effort required

Engineering Contradiction:

Improvestatistical accuracy through larger sample sizesVSAvoidcapacity to evaluate large quantities of production data

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual verification processes with automated image processing and OCR systems that can process large volumes of production data simultaneously, enabling statistical accuracy based on large sample sizes without the linear increase in human effort that would otherwise be required

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS8498948B2Method and system for assessing data classification quality

Publication Date: 2013.07.30 ADI INCORPORATED

AI summary

Production data classified from a data source, such as a plurality of handprinted forms, is compared to provisional truth data independently classified from the same data source for constructing master truth data. The production data is compared to the master truth data for evaluating the quality with which the production data was classified.