Knowledge Graph Process Mining for Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing process mining tools are limited in their ability to utilize additional meaningful data for tasks such as prediction and prescription of future events and activities, as they restrict input to only three basic attributes, failing to exploit the full potential of event data for applications like resource allocation and performance improvement.
Innovation Solution
The method involves computing a representation of events and process models as a common knowledge graph, leveraging semantic information from event logs to enable graph-based machine learning for enhanced process support, including prediction and prescription of future events and activities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If process mining tools use only three basic attributes (Case ID, Timestamp, Activity type), then the tools are simple and easy to operate, but they cannot exploit the full potential of event data for prediction and prescription tasks
Solution Approach 1:
The patent combines event logs, process models, and context information into a unified knowledge graph structure. This merging allows the system to leverage multiple data sources simultaneously, enabling prediction and prescription tasks while maintaining a consistent data representation framework.
Solution Approach 2:
The patent introduces a knowledge graph as an intermediary layer between raw event data and machine learning algorithms. This knowledge graph transforms diverse data sources into a standardized format with entities, attributes, and relationships, making the data suitable for graph-based machine learning without requiring complex preprocessing for each specific task.
2Measurement precision
If process mining tools integrate multiple data sources (event logs, process models, context information), then prediction and prescription accuracy improves, but the system complexity increases
Solution Approach 1:
The patent replaces traditional data processing mechanisms with graph-based machine learning approaches. Instead of using conventional statistical methods or sequential processing of multiple data sources, the system uses graph neural networks that naturally handle the structured relationships in the knowledge graph, simplifying the integration of multiple data sources while improving prediction accuracy.
3Loss of information
If additional context information is added to event data, then the meaningfulness and utility of the data increases, but the data processing and storage requirements increase
Solution Approach 1:
The patent extracts essential semantic information from context sources and represents it in a structured knowledge graph format. Instead of storing and processing entire context documents or unstructured information, the system extracts key entities, attributes, and relationships, retaining the meaningful semantic content while significantly reducing data volume and processing requirements.
Data Source
AI summary
A method for knowledge-based process support, and a process model is related to the process. The method includes providing an event log of event data related to process events by a data mining tool. The method also includes exploiting semantic information contained in the process events or event log by computing a representation of the events, the process model, and semantic information as a common knowledge graph, and using the knowledge graph in graph-based machine learning for the support of the process.