Semantic Aggregated Modeling for Textual Information Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current AI-based semantic systems focus on individual document modeling, lacking an integrated approach for aggregated modeling, search, visualization, and summarization across vast textual information, making it difficult to effectively identify and characterize relevant information in a semantically meaningful and condensed manner.
Innovation Solution
An integrated AI-based system and method for semantic aggregated modeling, search, visualization, and summarization, utilizing a semantic information server that connects various service users to provide services such as landscape studies, trend estimation, clearance evaluations, and patentability assessments by analyzing textual content from diverse sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If individual document modeling is used, then document-level semantic analysis is achieved, but aggregated modeling across vast textual information is not possible
Solution Approach 1:
The patent merges multiple individual document models into a unified aggregated model that represents a collection of documents. The system combines semantic information from multiple documents using weighting mechanisms and aggregation functions to create a holistic representation that captures both individual document characteristics and collective information patterns.
Solution Approach 2:
The patent creates a multi-functional system that can perform both individual document analysis and aggregated collection analysis. The modeling framework is designed to handle diverse information types and purposes, providing universal applicability across different information retrieval and analysis scenarios.
2Measurement precision
If comprehensive information analysis is performed, then characterization accuracy is improved, but information presentation becomes too detailed and loses condensation
Solution Approach 1:
The patent extracts essential semantic features and key characteristics from comprehensive information while discarding redundant details. The system identifies and retrieves only the most relevant information elements needed for accurate characterization, presenting condensed yet meaningful representations that maintain precision without overwhelming detail.
Solution Approach 2:
The patent transforms detailed information into condensed representations by changing the parameters of information presentation. The system adjusts the level of detail, aggregation granularity, and representation format to balance characterization accuracy with information condensation, allowing flexible presentation adapted to different user needs.
3Loss of information
If detailed semantic analysis is performed, then information understanding is improved, but processing time increases
Solution Approach 1:
The patent performs preliminary semantic analysis and pre-processing of information to prepare aggregated models in advance. By conducting initial characterization and organizing semantic structures beforehand, the system reduces the time required for detailed analysis when queries are executed, as much of the understanding work is already completed during model construction.
4Adaptability or versatility
If integrated modeling approach is implemented, then system capability is improved, but device complexity increases
Solution Approach 1:
The patent segments the integrated modeling system into distinct functional modules including document processing units, aggregation mechanisms, semantic analysis components, and presentation layers. This modular segmentation maintains high system capability while managing complexity through organized, independent components that can be developed and maintained separately.
Data Source
AI summary
The present teaching relates to method, system, medium, and implementations for modeling an entity based on textual information. When information about an entity to be modeled is received, textual information related to the entity is searched, including a plurality of documents related to at least one aspect of the entity. First aggregated semantic models are obtained for the plurality of documents. Each first aggregated semantic model represents one of the plurality of documents and includes a semantic feature vector and a semantic signature. Such first aggregated semantic models are clustered into at least one group, each of which corresponds to one of the at least one aspect of the entity and includes a set of first aggregated semantic models representing semantics of documents related to the aspect of the entity. A second aggregated semantic model is then derived based on each group of first aggregated semantic models to characterize one aspect of the entity, yielding one or more second aggregated semantic models. A second aggregated semantic model for an aspect of the entity comprises an aggregated semantic feature vector and an aggregated semantic signature for characterizing the aspect of the entity and facilitating a search with respect to the aspect.