Entity Relationship Vectorization for Multi-Person Text Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data processing techniques, such as tokenization, vectorization, and Conditional Random Fields (CRFs), fail to effectively organize facts in multi-person sentences, limiting their ability to identify appropriate entity relationships in natural language processing systems.
Innovation Solution
Implementing a system that analyzes textual content to identify entities, performs entity type classification, assigns entities to segments based on facts about people, recognizes relationships between entities, and converts this information into relationship vectors, allowing for autonomous operations in software applications and electronic devices to present relevant facts without user intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data processing techniques (tokenization, vectorization, CRFs) are used, then processing speed and simplicity are maintained, but the ability to organize facts in multi-person sentences and identify entity relationships deteriorates
Solution Approach 1:
The patent segments the multi-person sentence into multiple person-specific segments, each associated with a specific person entity. This segmentation allows the system to organize facts about each person separately while maintaining their relationships, thereby improving entity relationship identification accuracy without requiring a complete overhaul of the processing system.
Solution Approach 2:
The patent introduces a new dimensional structure by creating person-specific segments and relationship vectors that capture entity relationships across multiple persons. This dimensional expansion transforms the flat processing approach into a multi-layered structure that preserves both individual person facts and their interrelationships, improving reliability while managing complexity through structured organization.
2Measurement precision
If advanced machine learning classifiers are implemented to accurately organize facts and relationships, then entity relationship identification accuracy improves, but processing time and computational resources increase
Solution Approach 1:
By segmenting the text into person-specific portions and applying classifiers independently to each segment, the system achieves precise entity classification for each person while reducing the overall computational burden compared to processing the entire multi-person sentence as a single unit. This segmentation enables parallel processing and reduces the time complexity of classification operations.
Solution Approach 2:
The patent applies classification operations selectively to relevant segments containing entity mentions rather than processing the entire text uniformly. This partial action approach focuses computational resources on critical portions of the text, achieving high measurement precision for entity relationships while minimizing unnecessary processing time spent on irrelevant content.
3Loss of information
If facts are attributed to multiple individuals in multi-person sentences, then completeness of information organization improves, but difficulty of detecting and measuring relationships increases
Solution Approach 1:
The patent segments the multi-person sentence into person-specific segments, each containing facts attributed to a specific individual. This segmentation makes relationship detection more manageable by breaking down the complex multi-person context into smaller, more tractable units while preserving all factual information through systematic organization in relationship vectors.
Solution Approach 2:
The patent introduces relationship vectors as intermediary structures that mediate between the segmented person-specific facts and the overall relationship detection task. These vectors serve as a structured representation that captures entity relationships across multiple persons, making the detection and measurement of relationships more systematic and less complex by providing an intermediate layer of organization.
Data Source
AI summary
Systems and methods comprising: analyzing an electronic resource to identify Entities in textual content (wherein each Entity comprises word(s)); performing machine learning operations to assign an entity type classification of a plurality of entity type classifications to at least one of the Entities; performing machine learning operations to assign each said Entity to one or more segments of the textual content that respectively comprise facts about people; performing machine learning operations to recognize relationships of the Entities to each person or business entity identified in the textual content and assign a relationship classification of a plurality of relationship classifications to at least one of the Entities associated with one of the recognized relationships; converting the electronic resource into relationship vectors based on outputs of the first, second and third classifiers; and controlling operations of a software application using the relationship vector.