https://github.com/aarkue/ocpq-eval

https://github.com/aarkue/ocpq-eval

Last synced: 3 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/aarkue/ocpq-eval
• Owner: aarkue
• Created: 2025-01-10T21:33:47.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2025-01-24T13:52:10.000Z (over 1 year ago)
• Last Synced: 2025-01-24T14:35:01.938Z (over 1 year ago)
• Language: Jupyter Notebook
• Size: 1.41 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# OCPQ Evaluation Data

This repository holds evaluation data for the OCPQ tool.

Also see https://ocpq.aarkue.eu/ and https://github.com/aarkue/OCPQ.

The software version of OCPQ used is equivalent with the release v0.6.7, available at https://github.com/aarkue/OCPQ/releases/tag/v0.6.7.

We tested our implementation by running the backend web server in release mode.

The evaluation was performed on an OCEL 2.0 file based on the BPI Challenge 2017 (see https://data.4tu.nl/datasets/34c3f44b-3101-4ea9-8281-e38905c68b8d/1 and https://data.4tu.nl/datasets/5c9717a0-4c22-4b78-a3ad-d2234208bfd7/1 as well as https://data.4tu.nl/datasets/6889ca3f-97cf-459a-b630-3b0b0d8664b5/1).

The used files (OCEL 2.0 in JSON and SQLite format) are available for download as `./bpic2017-ocel2.zip` (OCEL 2.0 JSON), `./bpic2017.sqlite.zip`, and `./bpic2017.duckdb.zip` in this repository.

The Neo4j dump, is available [here](https://data.4tu.nl/datasets/5c9717a0-4c22-4b78-a3ad-d2234208bfd7/1).

## Results

![](./heatmap.svg)

Note, that a previous version of this repository used a different evaluation approach, and the SQLite durations where measured individually. In the current version, `sqlite3` from Python is used.

Moreover, the SQL queries were different, and updated ones (also compatible with DuckDB) were used for the newest evaluation results.

Below, the query/constraint formulations in the graphical approaches are included.

For SQL (SQLite) and Cypher (Neo4j), see the `sql.txt` and `neo4j-cypher.txt` files available in each subdirectory (Q1 - Q7).

The exported versions of the queries/constraints in OCPQ are also available (e.g., as `Q1.json` for usage in frontend, simply copy the text content and paste using alt+v in the OCPQ tool).

We also included all raw measurements as .json files, containing the measured durations in seconds.

## Q1

### DECLARE

![](./Q1/DECLARE.drawio.svg)

### OCBC

![](./Q1/OCBC.drawio.svg)

### **_OCPQ_**

![](./Q1/Q1.png)

## Q2

### DECLARE

![](./Q2/DECLARE.drawio.svg)

### OCCG

![](./Q2/OCCG.drawio.svg)

### OCBC

![](./Q2/OCBC.drawio.svg)

### **_OCPQ_**

![](./Q2/Q2.png)

## Q3

### OCCG

![](./Q3/OCCG.drawio.svg)

### OCBC

![](./Q3/OCBC.drawio.svg)

### **_OCPQ_**

![](./Q3/Q3.png)

## Q4

### OCBC

![](./Q4/OCBC.drawio.svg)

### **_OCPQ_**

![](./Q4/Q4.png)

## Q5

### **_OCPQ_**

![](./Q5/Q5.png)

## Q6

### **_OCPQ_**

![](./Q6/Q6.png)

## Q7

### **_OCPQ_**

![](./Q7/Q7.png)