Knowledge Profiles for Keyword-Independent Search Accuracy

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current textual searching methods rely on keywords, leading to inaccuracies and inefficiencies as they fail to retrieve relevant documents if the wrong keywords are used, resulting in missed information and irrelevant hits.

Innovation Solution

The development of methods and systems for generating, editing, and searching with knowledge profiles that utilize textual analysis and knowledge discovery, including homograph disambiguation, context profiling, and knowledge network construction, to identify and retrieve relevant information independently of keyword usage.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If keyword-based textual searching methods are used, then the search process is simple and fast, but the search accuracy decreases and relevant documents are missed

Engineering Contradiction:

Improvesearch accuracyVSAvoidsearching method complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent introduces knowledge profiles as an intermediary between the search query and the document collection. These profiles contain structured knowledge about documents (concepts, entities, relationships) that mediate the matching process, enabling more accurate retrieval without requiring complex keyword analysis of every document

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system performs preliminary analysis to create knowledge profiles for documents before the actual search occurs. This pre-processing extracts and structures key information (concepts, entities, relationships) in advance, so that during search, the system can quickly match queries against these pre-computed profiles rather than analyzing raw text in real-time

Inventive Principle:

Principle #10Preliminary action

2Productivity

If keyword-based searching is used, then the searching method is simple to implement, but the time to retrieve relevant information increases due to irrelevant hits

Engineering Contradiction:

Improveinformation retrieval efficiencyVSAvoidsearch time

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

Knowledge profiles serve as intermediaries that pre-structure document information into searchable concepts and entities. This mediation allows the system to quickly filter and rank documents based on profile matches rather than scanning full text, significantly reducing search time and eliminating irrelevant hits

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The patent segments documents into discrete knowledge units (concepts, entities, relationships) within knowledge profiles. This segmentation transforms unstructured text into organized, queryable elements, enabling efficient retrieval by matching specific knowledge units rather than processing entire documents

Inventive Principle:

Principle #1Segmentation

3Measurement precision

If truncation is used to broaden search terms, then more documents are retrieved, but the number of irrelevant hits increases

Engineering Contradiction:

Improvesearch result relevanceVSAvoidnumber of search results

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The system applies different levels of specificity to different parts of the search process. Knowledge profiles contain both broad contextual information and specific detailed concepts, allowing the search to retrieve comprehensive results while maintaining precision through localized matching of specific knowledge units within the profiles

Inventive Principle:

Principle #3Local quality

Data Source

PatentUS9971974B2Methods and systems for knowledge discovery

Publication Date: 2018.05.15 ELSEVIER INC

AI summary

Provided are methods and systems for knowledge discovery utilizing knowledge profiles.