Term Granularity Division via Parsing Element Words
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for dividing compound nouns, such as those used in financial and business systems, are inefficient as they rely on morphological analysis and practitioner judgment, leading to inconsistent results and the need for extensive dictionaries, which are impractical for large-scale systems.
Innovation Solution
A method that extracts element words from content using parsing techniques to divide terms at specific positions, allowing for dynamic granularity without relying on predefined lists or practitioner interpretation, using techniques like dependency parsing and predefined character separation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If morphological analysis technique is used to divide compound nouns, then division can be performed automatically, but the division result does not necessarily match the desired granularity because it relies on system dictionary and grammar
Solution Approach 1:
The patent introduces an intermediary process between automatic morphological analysis and final term division. A glossary management system acts as a mediator that collects division results, allows practitioner feedback, and iteratively improves the division algorithm. This intermediary layer enables the system to learn from actual usage patterns and refine its division accuracy over time without requiring complete re-engineering of the automatic analysis process.
Solution Approach 2:
The patent performs preliminary manual division of compound nouns by practitioners before system-wide deployment. These pre-divided terms serve as training data or reference standards that guide the automatic division algorithm. By establishing a foundation of correctly divided terms in advance, the system can achieve better initial accuracy and faster convergence during iterative improvement.
2Measurement precision
If manual division of compound nouns is performed by practitioners, then division accuracy can be improved, but it requires extensive time and cannot scale to large numbers of terms
Solution Approach 1:
The patent applies partial manual action only to critical or ambiguous terms rather than all compound nouns. The system identifies which terms require human judgment based on complexity metrics or frequency of use, and applies manual division selectively to those cases. This partial approach maintains high accuracy for important terms while preserving overall productivity by automating the majority of routine divisions.
Solution Approach 2:
The system enables practitioners to self-correct division results through a feedback mechanism. When automatic or preliminary manual division produces suboptimal results, practitioners can easily review and correct specific terms, and these corrections automatically become part of the training data. This self-service approach allows continuous improvement without requiring extensive external manual intervention for each new term.
3Measurement precision
If a comprehensive dictionary is created to cover all compound nouns, then division accuracy can be maintained, but the dictionary becomes unmanageably large and impractical for large-scale systems
Solution Approach 1:
The patent segments the comprehensive dictionary into multiple specialized sub-dictionaries organized by domain, term frequency, and division complexity. Instead of maintaining one massive dictionary, the system creates modular sub-dictionaries that can be independently managed and updated. This segmentation reduces the cognitive and computational burden while maintaining comprehensive coverage through organized categorization of terms.
Solution Approach 2:
The patent transforms the static comprehensive dictionary into a dynamic, evolving structure. The dictionary is no longer a fixed resource but grows and adapts continuously through automated collection of new terms and iterative refinement from practitioner feedback. This dynamic approach allows the system to handle large-scale terminology evolution without requiring manual reorganization of the entire dictionary structure.
4Adaptability or versatility
If practitioner judgment is used to determine main terms and modifiers, then context-appropriate division can be achieved, but the results become inconsistent across different practitioners
Solution Approach 1:
The patent implements a structured feedback loop where practitioner division decisions are collected, analyzed, and used to refine the automatic division algorithm. Practitioners provide feedback on ambiguous cases, and this feedback is systematically processed to update the glossary and improve future automatic divisions. This feedback mechanism gradually aligns practitioner judgments with algorithmic decisions, reducing inconsistency over time.
Solution Approach 2:
The patent changes the parameters of the division process from purely judgment-based to a hybrid model that incorporates statistical analysis of term usage patterns, collocations, and contextual information. By introducing these quantifiable parameters, the system reduces reliance on subjective practitioner interpretation while maintaining context-appropriateness through data-driven insights into how terms are actually used.
Data Source
AI summary
A method, computer system, and computer program product for dividing a term with appropriate granularity includes extracting an element word specifying granularity from content by parsing, and, if the term includes at least one element word in a part thereof, dividing the term at a position where the at least one element word exists.