This repository collects queries for modeling, importing, and analyzing event data as Knowledge Graphs using the Labeled Property Graph data model of graph databases. All scripts and queries are licensed under LGPL v3.0, see LICENSE. Copyright information is provided within each Project.
The following projects are part of this repository
Data model and generic query templates for importing, integrating, and transforming a set of related CSV event logs into single event knowledge graph (EKG), stored as labeled property graph in Neo4J. See csv_to_eventgraph_neo4j/README.txt
Publications:
- Stefan Esser, Dirk Fahland: Multi-Dimensional Event Data in Graph Databases. CoRR abs/2005.14552, Journal on Data Semantics, DOI: 10.1007/s13740-021-00122-1 (2020)
- Esser, Stefan. (2020, February 19). A Schema Framework for Graph Event Data. Master thesis. Eindhoven University of Technology. https://doi.org/10.5281/zenodo.3820037
Adaptation of the data model and query templates for import CSV event logs into an event knowledge graph (EKG) uzing KuzuDB.
- Includes experiments for object-centric process querying (OCPQ) reaching state-of-the-art performance using an off-the-shelf graph data managemente system, see OCPQ experiment results
- See csv_to_eventgraph_kuzudb/README.md
Implementation of an explorative case study to model the BPI Challenge 2017 data sets using Neo4J. Publications:
- Capita Selecta report over "Using Graph Data Structures for Event Logs"
- Stefan Esser, Dirk Fahland: Storing and Querying Multi-dimensional Process Event Logs Using Graph Databases. Business Process Management Workshops 2019: 632-644