Association Rule Mining via Attribute Change Tracking

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Analyzing large data sets to determine association rules is inefficient, as it generates many rules, with many being non-useful, and it is time-consuming to identify relevant ones, especially in specialized fields.

Innovation Solution

A machine learning technique that determines association rules by tracking changes to attribute values of data objects, using changed attribute values as consequents and other attribute values as antecedents, with constraints on scope attributes and occurrence frequency thresholds, to automatically implement relevant rules.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional data mining methods are used to analyze large data sets, then comprehensive association rules can be generated, but the analysis becomes time-consuming and generates many non-useful rules

Engineering Contradiction:

Improvequality of association rulesVSAvoidtime to identify relevant rules

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary actions by tracking attribute value changes before full analysis. Change logs capture attribute modifications in advance, allowing the system to pre-identify potential association rules based on actual data dynamics rather than analyzing entire datasets from scratch, thus reducing time to identify relevant rules while maintaining quality

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The system changes parameters by focusing analysis on changed attribute values rather than all attributes. By monitoring which attributes have changed and using those changes as the basis for rule generation, the system reduces the search space significantly, eliminating many non-useful rules while preserving reliable associations

Inventive Principle:

Principle #35Parameter changes

2Adaptability or versatility

If all available data is analyzed to determine association rules, then complete rule sets are generated, but computing resources and human resources are excessively consumed

Engineering Contradiction:

Improvecoverage of association rulesVSAvoidcomputing and human resources

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The system extracts only the necessary subset of data for analysis by focusing on changed attribute values and their relationships. Instead of processing all available data, it extracts change log entries and related attribute information, significantly reducing computing and human resources while maintaining adaptability through targeted analysis of relevant data portions

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system segments the data analysis process into focused components: tracking specific attribute changes, identifying related attributes through change logs, and generating rules only for observed change patterns. This segmentation allows comprehensive rule coverage for changed attributes while minimizing resource consumption by excluding unchanged data from processing

Inventive Principle:

Principle #1Segmentation

3Adaptability or versatility

If association rules are determined without constraints on scope attributes, then more potential rules are generated, but many rules become non-probative and non-useful

Engineering Contradiction:

Improvenumber of association rulesVSAvoidprobative value of rules

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The system applies local quality by imposing different constraints on different attributes based on their roles. Scope attributes have specific constraints applied to them (such as occurrence frequency thresholds), while other attributes are analyzed in relation to changed attributes. This selective application of quality controls ensures probative value while maintaining appropriate rule diversity

Inventive Principle:

Principle #3Local quality

4Loss of information

If traditional methods analyze entire data sets, then all possible relationships are captured, but the process is inefficient and generates excessive non-useful rules

Engineering Contradiction:

Improvecompleteness of data relationshipsVSAvoidefficiency of rule determination

Core Design Contradiction:

Loss of informationVSProductivity

Solution Approach 1:

The system performs preliminary tracking of attribute changes to capture relationship information before full analysis. Change logs record attribute modifications in advance, preserving complete relationship information for changed attributes while enabling efficient processing by pre-organizing data according to actual changes rather than requiring complete dataset re-analysis

Inventive Principle:

Principle #10Preliminary action

Data Source

PatentUS11556838B2Efficient data relationship mining using machine learning

Publication Date: 2023.01.17 SAP SE

AI summary

Techniques and solutions are described for determining association rules using a machine learning technique on a subset of data to which the association rules might apply, and from which they can be determined. In particular, association rules are determined by tracking changes to attribute values of data objects having a type. The changed attribute value can be used as a consequent in an association rule. Values of other attributes of data objects having the changed attribute value can be used as antecedents in association rules. Values used in antecedents can be constrained, such as by limiting values to those associated with scope attributes or values satisfying a threshold occurrence frequency. In some cases, determined association rules can be automatically implemented, such as to process input or stored data for data objects of the type.