Dynamic Ontology for Adaptive Data Discovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ontological systems are static and cannot evolve over time, leading to scalability and subjectivity issues in identifying new entities and relationships within large datasets, requiring manual intervention by data analysts and resulting in inconsistent classification standards and quality.
Innovation Solution
A dynamic ontology system that uses processors to identify data items in unstructured content, store unrecognized items with labels, generate weights, and update the schema when the weight exceeds a threshold, allowing for automated and adaptive data extraction and classification.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual ontology updates by data analysts are used, then ontology accuracy can be maintained, but scalability deteriorates due to difficulty in quickly identifying new entities and relationships within thousands of documents
Solution Approach 1:
The system enables self-service by automatically updating the ontology through machine learning models that identify new entities and relationships from unstructured documents without requiring manual analyst intervention. The ontology evolves autonomously by processing documents and incorporating newly discovered patterns, thereby maintaining accuracy while achieving scalability.
Solution Approach 2:
The patent replaces the mechanical manual process of ontology updates by data analysts with an automated computational system using machine learning and natural language processing. This substitution eliminates the bottleneck of human analysts reviewing thousands of documents, enabling the system to scale to large document volumes while maintaining or improving classification accuracy through consistent algorithmic application.
2Measurement precision
If multiple data analysts manually review documents to update ontology, then comprehensive entity identification is achieved, but subjectivity increases leading to varying classification standards and quality
Solution Approach 1:
The system applies homogeneity by using a unified machine learning model and standardized processing pipeline to identify entities and relationships across all documents. This ensures consistent classification standards are applied uniformly, eliminating the subjectivity and variability that arise when different human analysts review documents. The same algorithms and criteria are consistently applied throughout the system.
3Device complexity
If static ontology is used, then system simplicity is maintained, but adaptability deteriorates as the ontology cannot evolve through time to incorporate new entities and relationships
Solution Approach 1:
The patent implements dynamics by transforming the static ontology into a dynamic, evolving structure. The ontology automatically adapts to new information by incorporating newly discovered entities and relationships from processed documents. This dynamic update mechanism allows the system to remain simple in its core architecture while gaining the adaptability to evolve over time through automated learning from data.
Data Source
AI summary
A method, apparatus, system, and computer program code for intelligent data discovery with dynamic ontology are provided. According to one illustrative embodiment, the method using a number of processors to perform the steps of: identifying a set of data items in unstructured content using a dynamic data schema populated from a dynamic ontology; and responsive to identifying a data item that is not recognized in the data schema: storing the data item with labels; generating a weight for the data item; and responsive to the weight exceeding a threshold, updating the schema to include the data item that was not recognized.