Cangjie Code Simulation for Chinese Comprehension Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional language systems for Chinese communication face challenges in keeping pace with the evolving Chinese language, leading to slow and resource-intensive semantic searches, and difficulties in obtaining the latest usage trends due to the vast and dynamic nature of the language, which complicates building sufficient databases.

Innovation Solution

A Chinese common sense comprehension system that simulates Cangjie codes into digital codes, using a simulation module to classify and integrate concept information, and an integration module to convert these codes into program codes, thereby reducing the need for extensive databases and improving comprehension accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If conventional language systems use built-in databases or network search to process Chinese language, then they can provide response content, but the system becomes slow and consumes excessive hardware resources

Engineering Contradiction:
Improvelanguage processing speedVSAvoidhardware resource consumption
Core Design Contradiction:
ProductivityVSUse of energy by moving object

Solution Approach 1:

The patent segments Chinese characters into basic components (radicals and strokes) and represents them using Cangjie code structures. Each character is broken down into constituent parts that can be independently coded and recombined, transforming the processing approach from whole-character database matching to component-based algorithmic generation, thereby reducing resource consumption and improving processing speed

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent replaces the mechanical database search and network query systems with an algorithmic code simulation approach. Instead of physically searching through stored language data, the system uses computational algorithms to simulate Cangjie coding rules and generate meaning representations directly, substituting physical information retrieval with computational transformation

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Loss of information

If conventional systems search the Internet for semantic comparisons, then they can obtain language data, but the process is slow and cannot obtain the latest Chinese usage

Engineering Contradiction:
Improvelatest language usageVSAvoidsearch time
Core Design Contradiction:
Loss of informationVSLoss of time

Solution Approach 1:

The patent incorporates predefined Cangjie code structures and classification rules that encode fundamental Chinese language semantics in advance. These pre-encoded knowledge structures allow the system to immediately process and generate meanings for new or evolving language usage without requiring time-consuming external searches, as the foundational semantic framework is already embedded in the coding system

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent enables dynamic adaptation to evolving Chinese language by allowing the code simulation system to process new character formations and combinations as they emerge. The algorithmic approach can accommodate language evolution by applying consistent coding rules to new structures, maintaining up-to-date language understanding without relying on periodic database updates or web searches

Inventive Principle:
Principle #15Dynamics

3Adaptability or versatility

If the system builds comprehensive databases to cover 3,000 years of Chinese language evolution, then it can achieve complete language coverage, but the database becomes too large to effectively build and maintain

Engineering Contradiction:
Improvelanguage coverageVSAvoiddatabase size
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent decomposes the overwhelming task of storing all Chinese language data into manageable Cangjie code components. Instead of storing entire characters and their meanings in a massive database, the system stores fundamental coding rules and basic component definitions, then generates specific character meanings through algorithmic combination of these segments, dramatically reducing storage requirements while maintaining comprehensive language coverage

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a universal Cangjie coding framework that can handle diverse Chinese language elements across different historical periods and usage contexts through a single unified system. The same coding rules and structural principles apply whether processing ancient classical Chinese or modern colloquial expressions, eliminating the need for separate specialized databases for different language eras or registers

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentUS10846480B2Common sense comprehension system and method for comprehending Chinese common sense
Publication Date: 2020.11.24 CULTURE COM TECH MACAU
  • US10846480B2 patent drawing
  • US10846480B2 patent drawing
  • US10846480B2 patent drawing

AI summary

A Chinese common sense comprehension system includes a simulation module for simulating the Cangjie codes into concept information and an integration module for integrating the concept information into target information. Therefore, the Chinese common sense comprehension system adopts an innovative logical way of learning Chinese, thereby improving the accuracy of the artificial intelligence device to understand Chinese.