EA ModelSet

This repository features a dataset of EA models and the supporting applications to transform, analyze, and query the dataset.

Dataset

The dataset currently features 863 different EA models in .json, .xml (The Open Group Standard ArchiMate Model Exchange File Format), .archimate (Archi-specific XML model storage format) and .csv format.

Download & Usage

• Download the latest version of the dataset as a .zip archive
• Website with the dataset's landing page, search functionality, and statistics.
• Python Library to access the dataset in a Python environment

Schema & Structure

EA ModelSet follows a defined structure and leverages JSON schemas to provide a FAIR dataset of EA models. The main directory for the dataset is the dataset/ directory, which contains the dataset.json and processed-models/ sub-directory.

Dataset Directory Structure:

dataset/
|    dataset.json
+--- processed-models/
	+--- <id>/
	+--- <id>/
	...

The dataset.json files adheres to the ea-modelset.schema.json JSON schema and contains the dataset's metadata and an array of ModelInfo objects that provide a reduced subset of metadata and computed properties of each individual model.

Within the processed-models/ directory, each processed model has its own sub-directory with the model’s ID serving as the name.

Model Directory Structure:

processed-models/
+--- <model-id>/
    | model.archimate
    | model.json
    | model.xml
    +--- csv/
        | elements.csv
        | relations.csv
        | properties.csv
+--- <model-id>/
        ...

A model directory contains the primary JSON model file (model.json adhering to ea-model.schema.json), two ArchiMate XML model files (model.archimate and model.xml) and a csv/ directory with a model’s elements, relationships, and views stored as separate CSV files.

Contributions

TODO