Ontology Expansion via Entity-Association Rules

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current ontology programming methods are inadequate in processing and analyzing communication data across various domains, as they fail to adapt effectively to linguistic variations and do not efficiently extract meaningful patterns and relationships from large datasets.

Innovation Solution

The development of an ontology expansion method that processes communication data to extract significant phrases and phrase pairs, creating new abstract relations and relation instances, which are integrated into an initial ontology framework using machine learning techniques to enhance knowledge extraction and analysis.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If traditional ontology programming methods are used, then the ontology structure remains simple and manageable, but the ontology cannot effectively adapt to linguistic variations and extract meaningful patterns from communication data

Engineering Contradiction:

Improveadaptability to linguistic variationsVSAvoidontology expansion complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system performs self-training by automatically extracting significant phrases and phrase pairs from communication data, generating new abstract relations and relation instances without requiring manual ontology expansion. The machine learning model autonomously identifies patterns and expands the ontology structure based on observed linguistic variations in the training data.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The ontology is pre-trained using a training set of communication data before deployment. Significant phrases and phrase pairs are extracted in advance, and new abstract relations are created during the training phase, allowing the ontology to be ready for effective pattern detection when processing actual communication data.

Inventive Principle:

Principle #10Preliminary action

2Measurement precision

If manual ontology expansion is performed to capture domain-specific knowledge, then the ontology becomes more accurate for specific domains, but the process becomes time-consuming and labor-intensive

Engineering Contradiction:

Improvedomain knowledge accuracyVSAvoidontology development time

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The manual mechanical process of ontology expansion is replaced with an automated machine learning system. The system uses algorithms to extract significant phrases, compute prominence scores, identify phrase co-occurrences, and generate abstract relations automatically, substituting human effort with computational processes that are both faster and scalable.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

A machine learning model serves as an intermediary between raw communication data and the ontology structure. The model processes communication data to extract meaningful patterns and translates them into ontology elements (significant phrases, abstract relations, relation instances), bridging the gap between unstructured data and structured knowledge representation.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Adaptability or versatility

If the ontology is expanded to include more terms and relations to cover various domains, then the ontology becomes more versatile, but the processing and analysis complexity increases

Engineering Contradiction:

Improvedomain coverageVSAvoidpattern detection complexity

Core Design Contradiction:

Adaptability or versatilityVSDifficulty of detecting and measuring

Solution Approach 1:

The system changes parameters by computing prominence scores for n-grams and using statistical measures (log-likelihood ratios) to identify significant phrase co-occurrences. These parameter-based filtering mechanisms allow the ontology to focus on statistically significant patterns rather than all possible relations, managing complexity through quantitative thresholds.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The ontology expansion process is segmented into distinct stages: extracting significant phrases, identifying phrase pairs, computing prominence scores, detecting co-occurrences, and generating abstract relations. This segmentation allows each step to be optimized independently and manages overall complexity by breaking down the comprehensive task into manageable sub-tasks.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS11663411B2Ontology expansion using entity-association rules and abstract relations

Publication Date: 2023.05.30 VERINT SYST LTD

AI summary

A method for expanding an initial ontology via processing of communication data, wherein the initial ontology is a structural representation of language elements comprising a set of entities, a set of terms, a set of term-entity associations, a set of entity-association rules, a set of abstract relations, and a set of relation instances. A method for extracting a set of significant phrases and a set of significant phrase co-occurrences from an input set of documents further includes utilizing the terms to identify relations within the training set of communication data, wherein a relation is a pair of terms that appear in proximity to one another.