ComGen_CBCMS_plus: Artifact for Compliance Policy Generation

Fuente: Zenodo
Guardado en:
Detalles Bibliográficos
Autores principales: Zhuang, Zhixian, Lee, Xiaodong, Zhang, Aiyao, Wei, Jiuqi, Fu, Yufan, Peng, Botao
Formato: Recurso digital
Publicado: Zenodo 2025
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866901097196552192
author Zhuang, Zhixian
Lee, Xiaodong
Zhang, Aiyao
Wei, Jiuqi
Fu, Yufan
Peng, Botao
author_facet Zhuang, Zhixian
Lee, Xiaodong
Zhang, Aiyao
Wei, Jiuqi
Fu, Yufan
Peng, Botao
contents <h2>Artifact Overview</h2> <p>This release <strong>v1.0.0</strong> provides a frozen snapshot of the implementation used in the manuscript <em>under review</em>:</p> <p><strong>Structured Policy Modeling and Context-Aware Generation for Multi-Jurisdictional Compliance in Global Software Systems</strong><br> submitted to <em>Information and Software Technology (IST)</em>.</p> <p>The artifact corresponds to the version used for all experiments and evaluations reported in the manuscript.</p> <h2>Repository Description</h2> <p><strong>ComGen for CBCMS+ (Compliance Policy Generator)</strong></p> <p>This repository implements the <strong>Compliance Policy Generator (ComGen)</strong>, which constitutes the policy generation component of the CBCMS+ framework. ComGen adopts a Random Forest–based learning approach to generate compliance-related policies from structured input features, enabling efficient and scalable policy instantiation for global software systems operating under multiple regulatory regimes.</p> <h2>Key Features</h2> <ul> <li><strong>Structured Data Preprocessing:</strong> Transforms raw inputs into feature representations suitable for model training and inference.</li> <li><strong>Random Forest–Based Policy Generation:</strong> Predicts binary policy decisions across multiple outputs.</li> <li><strong>Hyperparameter Optimization:</strong> Employs 5-fold cross-validation for robust parameter selection.</li> <li><strong>Model Evaluation:</strong> Reports precision, recall, and F1-score metrics.</li> <li><strong>Flexible Training Modes:</strong> Supports both standard training and cross-validation–based workflows.</li> </ul> <h2>Contents of This Release</h2> <ul> <li>Core implementation of the ComGen pipeline</li> <li>Data preprocessing modules</li> <li>Model training and evaluation scripts</li> <li>Documentation and usage instructions in <code>README.md</code></li> </ul> <h2>Usage</h2> <ol> <li>Clone this repository at tag <code>v1.0.0</code>.</li> <li>Follow the instructions in <code>README.md</code> to install dependencies and prepare the environment.</li> <li>Execute the provided scripts to reproduce the policy generation and evaluation results reported in the manuscript.</li> </ol> <h2>Relation to the Manuscript</h2> <p>This artifact is released to support transparency and reproducibility during the peer-review process.<br> It reflects the exact implementation used in the submitted version of the manuscript and does not introduce functionality beyond what is evaluated and discussed in the paper.</p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_18016703
institution Zenodo
language
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle ComGen_CBCMS_plus: Artifact for Compliance Policy Generation
Zhuang, Zhixian
Lee, Xiaodong
Zhang, Aiyao
Wei, Jiuqi
Fu, Yufan
Peng, Botao
<h2>Artifact Overview</h2> <p>This release <strong>v1.0.0</strong> provides a frozen snapshot of the implementation used in the manuscript <em>under review</em>:</p> <p><strong>Structured Policy Modeling and Context-Aware Generation for Multi-Jurisdictional Compliance in Global Software Systems</strong><br> submitted to <em>Information and Software Technology (IST)</em>.</p> <p>The artifact corresponds to the version used for all experiments and evaluations reported in the manuscript.</p> <h2>Repository Description</h2> <p><strong>ComGen for CBCMS+ (Compliance Policy Generator)</strong></p> <p>This repository implements the <strong>Compliance Policy Generator (ComGen)</strong>, which constitutes the policy generation component of the CBCMS+ framework. ComGen adopts a Random Forest–based learning approach to generate compliance-related policies from structured input features, enabling efficient and scalable policy instantiation for global software systems operating under multiple regulatory regimes.</p> <h2>Key Features</h2> <ul> <li><strong>Structured Data Preprocessing:</strong> Transforms raw inputs into feature representations suitable for model training and inference.</li> <li><strong>Random Forest–Based Policy Generation:</strong> Predicts binary policy decisions across multiple outputs.</li> <li><strong>Hyperparameter Optimization:</strong> Employs 5-fold cross-validation for robust parameter selection.</li> <li><strong>Model Evaluation:</strong> Reports precision, recall, and F1-score metrics.</li> <li><strong>Flexible Training Modes:</strong> Supports both standard training and cross-validation–based workflows.</li> </ul> <h2>Contents of This Release</h2> <ul> <li>Core implementation of the ComGen pipeline</li> <li>Data preprocessing modules</li> <li>Model training and evaluation scripts</li> <li>Documentation and usage instructions in <code>README.md</code></li> </ul> <h2>Usage</h2> <ol> <li>Clone this repository at tag <code>v1.0.0</code>.</li> <li>Follow the instructions in <code>README.md</code> to install dependencies and prepare the environment.</li> <li>Execute the provided scripts to reproduce the policy generation and evaluation results reported in the manuscript.</li> </ol> <h2>Relation to the Manuscript</h2> <p>This artifact is released to support transparency and reproducibility during the peer-review process.<br> It reflects the exact implementation used in the submitted version of the manuscript and does not introduce functionality beyond what is evaluated and discussed in the paper.</p>
title ComGen_CBCMS_plus: Artifact for Compliance Policy Generation
url https://doi.org/10.5281/zenodo.18016703