Association Analysis Noise Reduction via Data Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Association analysis in transaction data is hindered by noise, leading to a large number of association rules that slow down the data association analyzer's ability to understand customer purchasing behavior effectively.
Innovation Solution
A method that generates an association model with a support value greater than a minimum threshold, identifies and reduces the number of association rules by inverting data and comparing certainty factor values to delete noisy data, thereby updating the transaction data to process fewer rules efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If association analysis is performed on original transaction data without filtering, then comprehensive purchasing behavior patterns are captured, but the number of association rules becomes excessively large, slowing down analysis speed
Solution Approach 1:
The patent applies preliminary action by performing data inversion and certainty factor analysis before generating the final association rules. The system pre-processes the transaction data by creating inverted datasets and calculating certainty factors for potential rules, then uses this pre-computed information to filter out noisy data before the main association rule generation. This preliminary filtering reduces the volume of data that needs to be processed in the main analysis phase, thereby improving analysis speed while maintaining pattern completeness.
2Productivity
If noise reduction techniques are applied to filter transaction data, then analysis speed increases, but some valid purchasing patterns may be lost
Solution Approach 1:
The patent implements feedback by using certainty factor calculations to evaluate and refine the filtering process. The system generates candidate association rules, calculates their certainty factors based on the inverted transaction data, and uses this feedback to determine whether to retain or remove each rule. This feedback mechanism ensures that only rules with sufficiently high certainty (indicating valid patterns) are kept, while filtering out noisy rules, thus preventing loss of valid purchasing patterns while maintaining analysis speed.
3Loss of time
If the number of association rules is reduced through filtering, then processing time decreases, but the quality of business insights may deteriorate
Solution Approach 1:
The patent applies parameter changes by transforming the transaction data into inverted form and using certainty factor thresholds as filtering parameters. The system changes the state of the data from original transactions to inverted transactions, and applies a certainty factor parameter threshold to filter rules. This parameter-based filtering reduces processing time by eliminating low-quality rules while maintaining the reliability of business insights through the mathematically sound certainty factor metric that preserves genuinely useful patterns.
Data Source
AI summary
Reducing noise during association analysis is provided. An association model is generated having a support value with respect to original data greater than a predefined minimum support value threshold level. A number of association rules corresponding to the association model are identified. It is determined whether the number of association rules corresponding to the association model is less than a predefined maximum number of association rules. In response to determining that the number of association rules corresponding to the association model is greater than the predefined maximum number of association rules, inverted data of the original data is generated. Another association model is generated having the support value with respect to the original data and the inverted data greater than the predefined minimum support value threshold level.