Text Relevance Determination via Knowledge Element Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining text relevance, such as those used in search applications, face challenges in accuracy due to reliance on explicit character or word level information, which fails to capture deeper meaning, and feature learning at the word level struggles with accurate text understanding and matching.
Innovation Solution
The introduction of a knowledge base that associates texts with corresponding entities, allowing for the determination of text relevance based on entity relevance between the entities in the knowledge base, improving the accuracy of relevance determination through knowledge element level analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If text relevance determination is based on explicit character or word level information, then the processing is simple and fast, but the accuracy of relevance determination deteriorates due to inability to capture deeper meaning
Solution Approach 1:
The patent segments text analysis into multiple levels: character level, word level, and knowledge element level. By dividing the analysis into hierarchical segments, the system can process simple character/word information quickly while also performing deeper knowledge element analysis to improve accuracy, thus resolving the contradiction between processing simplicity and determination accuracy.
Solution Approach 2:
The patent introduces knowledge elements as an intermediary between raw text and relevance determination. Knowledge elements serve as a bridge that captures deeper semantic meaning while maintaining structured analysis. This intermediary layer enables accurate relevance determination without requiring direct complex analysis of all text features, thus improving accuracy while managing complexity.
2Measurement precision
If feature learning is performed at the word level, then the processing is computationally efficient, but the text understanding and matching accuracy deteriorates
Solution Approach 1:
The patent segments feature learning into word-level features and knowledge element-level features. Word-level features provide computationally efficient processing, while knowledge element-level features provide deeper semantic understanding. This segmentation allows the system to balance computational efficiency with accurate text understanding by leveraging both levels of analysis.
Solution Approach 2:
The patent adds a new dimension to feature learning by introducing knowledge element features beyond traditional word-level features. This dimensional expansion from single-word features to knowledge element features enables richer semantic representation and more accurate text understanding without requiring exhaustive analysis of all possible word-level features, thus managing computational energy consumption.
Data Source
AI summary
Embodiments of the present disclosure provide a method, apparatus, device and medium for determining text relevance. The method for determining text relevance may include: identifying, from a predefined knowledge base, a first set of knowledge elements associated with a first text and a second set of knowledge elements associated with a second text. The knowledge base includes a knowledge representation consist of knowledge elements. The method may further include: determining knowledge element relevance between the first set of knowledge elements and the second set of knowledge elements, and determining text relevance between the second text and the first text based at least on the knowledge element relevance.