Self-learning data lenses for semantic conversion

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current machine-based tools for converting semantic information face challenges in handling context-dependent conversions, requiring significant time and expertise to establish knowledge bases and often struggle with ambiguity resolution due to rigid classification structures and limited context utilization.

Innovation Solution

A self-learning tool that statistically analyzes sample data to develop a knowledge base, leveraging existing conversion rules and statistical information to recognize valid attributes and attribute values, and uses frame-slot architectures to enhance term recognition and disambiguation, reducing the need for extensive training and improving accuracy.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a knowledge base is established using traditional machine-based tools, then conversion accuracy can be improved, but the time and expertise required increases significantly

Engineering Contradiction:

Improveconversion accuracyVSAvoidtime to establish knowledge base

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system performs self-learning by automatically analyzing training data to extract attributes, attribute values, and conversion rules without requiring manual configuration. The knowledge base is built autonomously through statistical analysis of sample data, eliminating the need for expert operators to manually establish conversion rules while maintaining high accuracy.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The system performs preliminary learning by analyzing training data in advance to build a knowledge base before actual conversion operations. This preliminary analysis extracts common patterns, attributes, and conversion rules that are then reused during conversion operations, reducing the time required for each conversion task.

Inventive Principle:

Principle #10Preliminary action

2Device complexity

If rigid classification structures are used for conversion, then system complexity is reduced, but the ability to handle context-dependent conversions deteriorates

Engineering Contradiction:

Improvesystem structure simplicityVSAvoidcontext-dependent conversion capability

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system uses dynamic attribute extraction where attributes and their valid values are not fixed in advance but are learned from training data and adapted to different contexts. The frame-slot architecture allows the system to dynamically identify relevant attributes based on the context of each conversion task, enabling flexible handling of context-dependent conversions while maintaining manageable system complexity.

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes parameters by learning different attribute sets and valid values for different contexts through statistical analysis of training data. Instead of using a fixed classification structure, the system adapts its parameters (attributes and values) based on the specific conversion context, allowing it to handle diverse conversion scenarios effectively.

Inventive Principle:

Principle #35Parameter changes

3Measurement precision

If extensive training data is used to improve accuracy, then conversion precision increases, but the resources and time required for training increase

Engineering Contradiction:

Improveconversion precisionVSAvoidtraining data volume

Core Design Contradiction:

Measurement precisionVSQuantity of substance

Solution Approach 1:

The system replaces manual configuration and expert knowledge with statistical analysis mechanisms. Instead of requiring experts to manually configure conversion rules based on extensive training, the system automatically analyzes training data using statistical methods to extract conversion patterns, reducing both the perceived training volume and the resources required for setup.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system identifies and copies common patterns and structures from training data to build reusable conversion rules. By detecting recurring attribute patterns and valid value sets in the training data, the system creates template-based conversion rules that can be applied across multiple conversion tasks, improving precision while reducing the effective training data volume needed.

Inventive Principle:

Principle #26Copying

Data Source

PatentUS9043367B2Self-learning data lenses for conversion of information from a first form to a second form

Publication Date: 2015.05.26 ORACLE INT CORP

AI summary

A semantic conversion system (1900) includes a self-learning tool (1902). The self-learning tool (1902) receives input files from legacy data systems (1904). The self-learning tool (1902) includes a conversion processor (1914) that can calculate probabilities associated with candidate conversion terms so as to select an appropriate conversion term. The self-learning tool (1902) provides a fully attributed and normalized data set (1908).