Context-Based Rules Engine for Explainable Content Relevance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing natural language processing systems face challenges in efficiently processing large volumes of complex digital content due to combinatorial complexity, resource intensity, and the need for substantial computing resources, while machine learning and deep learning methods require large annotated datasets and lack explainability, making human review resource-intensive and time-consuming.
Innovation Solution
A hybrid approach combining rules-based decision-making with machine learning and deep learning techniques to process natural language content, using context-based attributes and rules to determine relevance, enabling efficient and explainable responses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If machine learning and deep learning techniques are used to process natural language content, then automation and processing speed are improved, but large annotated datasets are required and explainability is lost
Solution Approach 1:
The patent segments the natural language processing task into multiple context-based attributes (CBA), where each attribute represents a specific aspect or dimension of the content. This segmentation allows the system to process complex content by breaking it down into manageable, independently analyzable components, reducing the need for large annotated datasets while maintaining processing effectiveness.
Solution Approach 2:
The patent transforms natural language content into a structured parameter space by evaluating multiple context-based attributes. Each attribute generates discrete outcome values that are combined to form a comprehensive representation of the content. This parameter transformation enables rule-based systems to process natural language efficiently without requiring extensive training data.
2Extent of automation
If machine learning and deep learning techniques are used to process natural language content, then automation is improved, but explainability of predictions deteriorates
Solution Approach 1:
The patent introduces context-based attributes as intermediary representations between natural language input and final predictions. These attributes serve as explicit, interpretable intermediaries that capture the rationale for predictions, enabling the system to maintain high automation levels while providing explainable outputs through the structured attribute evaluation process.
3Measurement precision
If a large number of rules are applied to natural language content using IF-THEN-ELSE statements, then processing accuracy is improved, but coding complexity and debugging time increase
Solution Approach 1:
The patent extracts the decision logic from complex IF-THEN-ELSE coding structures and relocates it into a configurable rules engine that operates on context-based attributes. This extraction separates the business logic from the implementation code, allowing high processing accuracy to be achieved through configurable rules rather than hard-coded conditional statements, thereby reducing coding complexity and debugging time.
4Speed
If natural language processing algorithms are allocated additional computing resources, then processing speed is improved, but resource intensity and cost increase
Solution Approach 1:
The patent implements a self-service processing architecture where the system automatically evaluates multiple context-based attributes and applies rules in a structured sequence without requiring additional computing resources. The inherent structure of the CBA approach enables efficient processing by organizing the evaluation logic to minimize redundant computations and resource consumption.
Data Source
AI summary
A decision support system for assessing and reviewing large volume of digital content which comprise complex subject matter and providing recommendations on relevance of each content by applying context-based rules which are specific to the subject matter of interest. The rules are captured in a standardized format and the algorithm for the rules-based decision making is designed with the flexibility to select the rules based on the subject matter of interest.