Weighted Fusion Classification for Data Confidence
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Low data quality significantly impacts the performance of machine learning algorithms, leading to increased errors in classification tasks, as it affects the confidence in automated decisions made by these systems.
Innovation Solution
A computer-implemented method and system that classify an input data set using multiple data recognition tools by identifying attributes, allocating confidence factors, and combining these factors through weighted fusion to enhance classification accuracy and confidence in data categorization, such as facial recognition or audio analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple data recognition tools are used to classify attributes, then classification accuracy and decision confidence are improved, but system complexity and computational resources increase
Solution Approach 1:
The system segments the classification task by identifying multiple distinct attributes of the data category (e.g., different facial features, audio characteristics) and processes each attribute separately through specialized data recognition tools, then combines the results through weighted fusion
Solution Approach 2:
The system merges the outputs of multiple data recognition tools by allocating confidence factors to each tool's classification and combining these through weighted fusion to produce a single output confidence, thereby improving overall classification accuracy while managing system complexity
2Reliability
If multiple data recognition tools are used with weighted fusion, then decision confidence is enhanced, but processing time and computational cost increase
Solution Approach 1:
The system performs preliminary identification of relevant attributes and selection of appropriate data recognition tools before processing, and pre-allocates confidence factors based on tool performance characteristics, thereby reducing processing time while maintaining high decision confidence
Solution Approach 2:
The system adjusts the weighting parameters in the weighted fusion based on the specific data category and attribute being classified, optimizing the balance between processing time and decision confidence for different scenarios
Data Source
AI summary
A computer-implemented method and system are disclosed for classifying an input data set within a data category using multiple data recognition tools. The method includes identifying at least a first attribute and a second attribute of the data category; classifying the at least first attribute via at least a first data recognition tool and the at least second attribute via at least a second data recognition tool, the classifying including: allocating a confidence factor for each of the at least first and second attributes that indicates a presence of each of the at least first and second attributes in the input data set; and combining outputs of the classifying into a single output confidence score by using a weighted fusion of the allocated confidence factors.