Machine Learning Rule Mining for Business Process Automation

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing computer-implemented processes face challenges in efficiently managing and updating complex business rules across various workflows, leading to increased time and costs due to the explosion of enterprise data.

Innovation Solution

The use of machine learning models, specifically association rule mining algorithms like K-Optimal Rule Discovery (KORD), to identify and optimize relevant business rules by scanning data sources, generating a list of rules, and displaying them in a graphical user interface for user acceptance or rejection, thereby reducing the computational resources required.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If traditional methods are used to manage and update business rules in complex computer-implemented processes, then comprehensive rule coverage can be achieved, but the time and costs for generating, updating, and managing business rules increase significantly

Engineering Contradiction:

Improvecomprehensive rule coverageVSAvoidtime for generating and updating business rules

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent replaces traditional mechanical/manual business rule management processes with machine learning-based automated rule mining. The system uses ML models to automatically discover, generate, and prioritize business rules from enterprise data, eliminating the need for manual rule creation and updating while maintaining comprehensive rule coverage across complex processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service rule mining where the machine learning model autonomously discovers business rules from data sources without requiring manual intervention. The model automatically processes enterprise data, identifies patterns, generates rules, and prioritizes them based on relevance, allowing the system to serve itself in rule generation and management.

Inventive Principle:

Principle #25Self-service

2Reliability

If traditional methods are used to manage and update business rules in complex computer-implemented processes, then comprehensive rule coverage can be achieved, but the costs for generating, updating, and managing business rules increase significantly

Engineering Contradiction:

Improvecomprehensive rule coverageVSAvoidcosts for generating and updating business rules

Core Design Contradiction:

ReliabilityVSEase of manufacture

Solution Approach 1:

The patent replaces expensive manual business rule management with automated machine learning-based rule mining. By using ML models to automatically discover and generate rules from enterprise data, the system eliminates labor costs associated with manual rule creation, review, and updates while maintaining comprehensive rule coverage across all business processes.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system creates copies of business rules by automatically generating multiple rule variations from the same underlying data patterns. The machine learning model can replicate successful rule patterns across different contexts and processes, reducing the need for manual rule creation and lowering overall management costs.

Inventive Principle:

Principle #26Copying

3Productivity

If machine learning models are used to rapidly identify and reuse rules, then computing resource consumption is reduced, but the complexity of the rule mining system increases

Engineering Contradiction:

Improverapid identification and reuse of rulesVSAvoidcomplexity of the rule mining system

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent introduces a machine learning model as an intermediary between enterprise data and business rule generation. The ML model serves as a mediator that automatically processes raw data, identifies patterns, and generates prioritized rule lists, simplifying the overall system architecture while enabling rapid rule identification and reuse across multiple business processes.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11783205B2Machine learning-based rule mining algorithm

Publication Date: 2023.10.10 SAP SE

AI summary

Data is received that defines a rule mining run including a scope of a search and at least one data source to be searched. In response, the at least one data source is polled to obtain rules responsive to the rule mining run. Each rule can specify one or more actions to take as part of a computer-implemented process when certain conditions are met. A list of rules (i.e., a proposed subset of the obtained rules) can then be generated using at least one machine learning model. The generated list of rule can then be displayed in a graphical user interface. Related apparatus, systems, techniques and articles are also described.