Hybrid Analytics Platform for Narrative Text Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current analytics platforms are inadequate in extracting meaningful information from narrative text data due to their reliance on 'context-thin' Natural Language Processing (NLP) methods that fail to scale, meet timeliness or efficiency objectives, and cannot derive meaning from text, leading to inefficiencies and limitations in data analysis.
Innovation Solution
A hybrid analytics platform that combines user configuration and machine learning to extract structured observations from narrative text, enabling flexible feed and data management, query-based alerting, scoring, and workflow management, and translating evolving user views into data models that represent objects, traits, and relationships in a meaningful way.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If NLP tools are used to analyze text data, then text processing capability is provided, but meaningful information extraction fails
Solution Approach 1:
The patent introduces a knowledge graph as an intermediary structure that bridges raw text data and meaningful information extraction. The knowledge graph serves as a mediator that organizes entities, relationships, and attributes in a structured format, enabling the system to extract meaningful information by traversing and querying this structured representation rather than processing raw text directly.
Solution Approach 2:
The patent segments the text analysis process into distinct stages: text processing, entity extraction, relationship identification, and knowledge graph construction. This segmentation allows each stage to focus on specific tasks, improving overall effectiveness in extracting meaningful information from text data.
2Productivity
If existing analytics platforms are used, then data analysis is provided, but scalability and timeliness objectives cannot be met
Solution Approach 1:
The patent implements dynamic knowledge graph updates that automatically reflect changes in real-time data streams. The system dynamically adjusts its analysis based on incoming data, enabling scalable processing of large volumes of data while maintaining timeliness through continuous updates rather than batch processing.
Solution Approach 2:
The system maintains continuous processing of data streams through persistent knowledge graph updates. Rather than periodic batch analysis, the knowledge graph is continuously updated as new data arrives, enabling real-time insights and meeting timeliness objectives while scaling with data volume.
3Productivity
If NLP methods are applied to text data, then processing is performed, but context understanding is insufficient
Solution Approach 1:
The patent transitions from flat NLP processing to a multi-dimensional knowledge graph structure that captures entities, relationships, and attributes across multiple levels. This dimensional expansion enables the system to understand context by traversing relationships between entities rather than relying solely on linear text processing.
Solution Approach 2:
The system combines multiple processing approaches (NLP for initial processing, knowledge graph construction for structural organization, and query processing for analysis) to create a composite analysis system that leverages the strengths of each component while mitigating their individual limitations.
Data Source
AI summary
An analytics platform for the extraction of structured observations from largely narrative sources using a hybrid approach of user configuration and machine learning is provided. The analytics platform collects and normalizes data from public and private sources and applies extractions to the data to create a world view of objects, traits, and relationships of interest and maintains that world view as data and/or extractions are updated. The platform is further configured to apply queries to the extracted world view for a variety of purposes including scoring objects for prioritized attention, generating notifications when specific conditions are met, providing data sets for exploratory analysis, and triggering the automatic collection of enhancing data from external sources.