ML Automated Entity Discovery System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current search engines rely heavily on user knowledge and manual query formulation, leading to suboptimal results due to variability in user analysis efficiency, and lack automation in discovering new entities relevant to a target category.
Innovation Solution
A machine-learning based automated search system that processes entity information documents using natural language processing and image analysis to identify search terms, execute searches across multiple data sources, and generate ranked lists of new entities, leveraging custom search APIs and similarity measures to enhance discovery and ranking accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual query formulation is used, then user control over search is maintained, but search accuracy and comprehensiveness deteriorate due to variability in user analysis efficiency
Solution Approach 1:
The system performs self-service by automatically analyzing entity information documents, extracting entities and attributes, and generating search queries without human intervention. The automated search system processes documents, identifies entities, and executes searches independently, eliminating the need for manual query formulation while maintaining accuracy.
Solution Approach 2:
The patent replaces the mechanical process of manual query formulation with an automated computational system. Instead of users manually analyzing documents and constructing queries, the system uses automated entity recognition, attribute extraction, and query generation algorithms to perform these tasks, substituting human cognitive work with computational processes.
2Productivity
If automated search is implemented, then productivity and comprehensiveness of entity discovery improve, but device complexity increases due to ML and NLP components
Solution Approach 1:
The automated search system performs multiple functions within a single integrated platform: document processing, entity recognition, attribute extraction, search query generation, and result ranking. This multi-functional approach consolidates what would otherwise require separate tools and processes, managing complexity through functional integration rather than proliferation of separate components.
Solution Approach 2:
The system utilizes parameter changes in machine learning models and natural language processing algorithms to adapt to different document types and search requirements. By adjusting parameters such as entity recognition thresholds, attribute weighting, and ranking criteria, the system maintains high productivity across diverse scenarios without requiring fundamentally different system architectures.
3Measurement precision
If comprehensive entity analysis is performed, then discovery accuracy improves, but loss of time increases due to detailed processing requirements
Solution Approach 1:
The system performs preliminary actions by pre-processing entity information documents to extract and store entity attributes in structured formats before actual search execution. Entity recognition and attribute extraction are performed in advance, creating ready-to-use data structures that can be quickly queried, thereby reducing processing time during actual search operations while maintaining comprehensive analysis accuracy.
Solution Approach 2:
The patent segments the entity analysis process into distinct modular stages: document preprocessing, entity recognition, attribute extraction, query generation, and result ranking. Each segment can be independently optimized and executed, allowing comprehensive analysis to be distributed across multiple processing steps rather than requiring monolithic processing, thus reducing overall processing time while maintaining accuracy.
Data Source
AI summary
A machine learning (ML) based automated search system receives an entity information document including a plurality of entities and identifies new entities that are similar to the plurality of entities which are not included in the entity information document via automated searches. Entity intelligence reports are generated for the plurality of entities which are further used to extract search terms. The search terms are used for executing automatic searches for documents with relevant portions. The documents are further analyzed to identify other, new entities which are not included in the entity information document. Entity intelligence reports including information regarding the new entities are also generated. Significant attributes are determined for the entities. The significant attributes are further employed in ranking the new entities so that top ranked new entities are identified. Alerts can be generated based on the information regarding the new entities.