Entity Relationship Extraction Workspace for Document Sensemaking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Knowledge workers face challenges in efficiently capturing and managing information from multiple documents, as existing technologies lack the ability to identify and record specific entities and relationships, and assign interest values, leading to time-consuming note-taking and limited computer-assisted analysis.

Innovation Solution

A system that allows knowledge workers to manipulate entities and relationships in a graphical user interface, enabling quick-click entity extraction, automatic relationship creation, and degree-of-interest assignment, facilitating efficient information management and analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If knowledge workers manually record entities and relationships in evidence files, then information can be captured, but the process becomes extremely time-consuming

Engineering Contradiction:

Improveinformation capture completenessVSAvoidnote-taking time

Core Design Contradiction:

Loss of informationVSLoss of time

Solution Approach 1:

The system enables self-service by automatically extracting entities and relationships from documents using computational algorithms, eliminating the need for manual information entry while maintaining comprehensive capture of relevant data

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces the mechanical manual note-taking process with automated computational text processing and natural language understanding algorithms that systematically identify and record entities and relationships

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Reliability

If knowledge workers use full text search to locate notes, then previously read documents can be re-found, but the computer cannot distinguish between text-snippets of interest and those that are not

Engineering Contradiction:

Improvedocument retrieval accuracyVSAvoidinformation filtering difficulty

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The system applies local quality by assigning different properties to different parts of the text, specifically marking entities and relationships with metadata tags that indicate their significance and type, enabling selective retrieval based on information quality rather than just text matching

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The patent uses visual differentiation (color coding and highlighting) to distinguish between different types of information elements, allowing users to quickly identify and retrieve specific entities and relationships of interest from large document collections

Inventive Principle:

Principle #32Color changes

3Quantity of substance

If knowledge workers hand-type notes about information, then captured information can be stored, but detailed note-taking becomes extremely time-consuming

Engineering Contradiction:

Improveinformation storage capacityVSAvoidinformation processing speed

Core Design Contradiction:

Quantity of substanceVSProductivity

Solution Approach 1:

The system creates automated copies of relevant information from source documents, extracting entities and relationships directly into structured formats without requiring manual transcription, thereby maintaining comprehensive information storage while dramatically improving processing speed

Inventive Principle:

Principle #26Copying

4Loss of information

If knowledge workers use traditional evidence files, then information can be recorded, but computer assistance remains limited without access to relationship information

Engineering Contradiction:

Improverelationship information captureVSAvoidsystem structure

Core Design Contradiction:

Loss of informationVSDevice complexity

Solution Approach 1:

The patent segments information into distinct structured components (entities, relationships, attributes) that can be independently processed and analyzed by the computer system, enabling sophisticated computer-assisted analysis while maintaining manageable system complexity through modular data organization

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS7668787B2Method, apparatus, and program product for efficiently adding entities and relationships in a comprehension state of a collection of information

Publication Date: 2010.02.23 GENESEE VALLEY INNOVATIONS LLC

AI summary

Aspects of the disclosed technology present a workspace window responsive to a relationship data structure that represents a comprehension state including a presentation set of an ordered set of text strings from an electronic document. The presentation set includes one or more identified strings. The workspace window can then receive a quick-click command invocation on the one or more identified strings and modifies the relationship data structure by adding an entity/relationship object to the relationship data structure responsive to the quick-click command invocation and the one or more identified strings. The application relates to sensemaking or maintaining a comprehension state of a document collection, by recording evidence, spatial hypertext, automatic highlighting, automating inferencing, reading recommendations, and reading through multiple documents.