SynthCity
Fuente:
Zenodo
Gespeichert in:
| 1. Verfasser: | |
|---|---|
| Format: | Recurso digital |
| Veröffentlicht: |
Zenodo
2026
|
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866902232244420608 |
|---|---|
| author | Agal, Sanjay |
| author_facet | Agal, Sanjay |
| contents | <h1>SynthCity Dataset</h1> <p><strong>A Multi-Modal Synthetic Dataset for Socially Realistic Smart City Systems</strong></p> <h2> Overview</h2> <p>The <strong>SynthCity dataset</strong> is a synthetic, multi-modal dataset generated using the <em>SynthCity framework</em>, designed to simulate realistic human behavior, mobility patterns, and social interactions in AI-driven smart city environments.</p> <p>This dataset enables researchers to evaluate machine learning, data science, and intelligent system models in scenarios where real-world data is limited due to privacy, accessibility, or ethical constraints.</p> <h2> Purpose</h2> <ul> <li> <p>Support <strong>reproducible research</strong> in smart city analytics</p> </li> <li> <p>Enable benchmarking of <strong>AI/ML models</strong> under realistic social conditions</p> </li> <li> <p>Provide <strong>privacy-preserving alternative</strong> to real-world datasets</p> </li> <li> <p>Facilitate research in:</p> <ul> <li> <p>Computational social systems</p> </li> <li> <p>Urban computing</p> </li> <li> <p>Intelligent transportation</p> </li> <li> <p>Resource allocation in 5G/6G systems</p> </li> </ul> </li> </ul> <h2> Dataset Structure</h2> <p>The dataset consists of the following components:</p> <h3>1. <code>population.csv</code></h3> <ul> <li> <p>Synthetic individuals with demographic attributes:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>age</code></p> </li> <li> <p><code>gender</code></p> </li> <li> <p><code>occupation</code></p> </li> <li> <p><code>income_level</code></p> </li> </ul> </li> </ul> <h3>2. <code>mobility.csv</code></h3> <ul> <li> <p>Simulated mobility patterns:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>timestamp</code></p> </li> <li> <p><code>location_id</code></p> </li> <li> <p><code>activity_type</code> (work, home, leisure, etc.)</p> </li> </ul> </li> </ul> <h3>3. <code>social_network.csv</code></h3> <ul> <li> <p>Social interaction graph:</p> <ul> <li> <p><code>source_agent</code></p> </li> <li> <p><code>target_agent</code></p> </li> <li> <p><code>interaction_weight</code></p> </li> </ul> </li> </ul> <h3>4. <code>resource_usage.csv</code></h3> <ul> <li> <p>Resource consumption patterns:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>timestamp</code></p> </li> <li> <p><code>resource_type</code> (energy, bandwidth, transport)</p> </li> <li> <p><code>usage_value</code></p> </li> </ul> </li> </ul> <h3>5. <code>environment.json</code></h3> <ul> <li> <p>Smart city configuration:</p> <ul> <li> <p>Zones</p> </li> <li> <p>Infrastructure nodes</p> </li> <li> <p>Simulation parameters</p> </li> </ul> </li> </ul> <h2>⚙️ Data Generation Methodology</h2> <p>The dataset is generated using a hybrid approach combining:</p> <ul> <li> <p><strong>Agent-Based Modeling (ABM)</strong> for simulating individual behavior</p> </li> <li> <p><strong>Graph-based modeling</strong> for social interactions</p> </li> <li> <p><strong>Generative AI techniques</strong> (e.g., probabilistic sampling / deep generative models)</p> </li> <li> <p><strong>Rule-based constraints</strong> to ensure realistic urban dynamics</p> </li> </ul> <h2> Key Features</h2> <ul> <li> <p>✔ Multi-modal (demographic, mobility, social, resource data)</p> </li> <li> <p>✔ Scalable (configurable population size)</p> </li> <li> <p>✔ Privacy-preserving (no real personal data)</p> </li> <li> <p>✔ Socially-aware behavior modeling</p> </li> <li> <p>✔ Suitable for benchmarking and simulation</p> </li> </ul> <h2> Potential Use Cases</h2> <ul> <li> <p>Smart city resource optimization</p> </li> <li> <p>Traffic and mobility prediction</p> </li> <li> <p>Social network analysis</p> </li> <li> <p>AI model benchmarking</p> </li> <li> <p>Policy simulation and decision support</p> </li> <li> <p>6G-enabled intelligent systems</p> </li> </ul> <h2> Data Size & Format</h2> <ul> <li> <p>Format: CSV / JSON</p> </li> <li> <p>Records: Configurable (default: 10,000 agents)</p> </li> <li> <p>Time span: Simulated over configurable time intervals</p> </li> </ul> <h2> Ethical Considerations</h2> <p>This dataset is <strong>fully synthetic</strong> and does not contain any real personal or sensitive data. It is designed to comply with ethical AI and data privacy standards.</p> <h2> License</h2> <p>This dataset is released under the <strong>Creative Commons Attribution 4.0 International (CC BY 4.0)</strong> license.</p> <h2> Citation</h2> <p>If you use this dataset, please cite:</p> <blockquote> <p>Agal, S. (2026). <em>SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems</em>. (Dataset)</p> </blockquote> <h2> Related Work</h2> <p>This dataset is associated with the research manuscript:</p> <p><strong>“SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems”</strong></p> <h2> Contact</h2> <p><strong>Dr. Sanjay Agal</strong><br>Professor & Head, Department of Artificial Intelligence and Data Science<br>Parul Institute of Engineering and Technology, India</p> <p>For queries: sanjay.agal32685@paruluniversity.ac.in</p> <h2> Future Work</h2> <p>Future releases will include:</p> <ul> <li> <p>Larger-scale simulations</p> </li> <li> <p>Real-time streaming data</p> </li> <li> <p>Integration with IoT sensor models</p> </li> <li> <p>Enhanced behavioral realism using advanced generative models</p> </li> </ul> <h2> Acknowledgment</h2> <p>We acknowledge the support of academic and research environments that facilitated the development of this dataset.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_19104742 |
| institution | Zenodo |
| language | |
| publishDate | 2026 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | SynthCity Agal, Sanjay <h1>SynthCity Dataset</h1> <p><strong>A Multi-Modal Synthetic Dataset for Socially Realistic Smart City Systems</strong></p> <h2> Overview</h2> <p>The <strong>SynthCity dataset</strong> is a synthetic, multi-modal dataset generated using the <em>SynthCity framework</em>, designed to simulate realistic human behavior, mobility patterns, and social interactions in AI-driven smart city environments.</p> <p>This dataset enables researchers to evaluate machine learning, data science, and intelligent system models in scenarios where real-world data is limited due to privacy, accessibility, or ethical constraints.</p> <h2> Purpose</h2> <ul> <li> <p>Support <strong>reproducible research</strong> in smart city analytics</p> </li> <li> <p>Enable benchmarking of <strong>AI/ML models</strong> under realistic social conditions</p> </li> <li> <p>Provide <strong>privacy-preserving alternative</strong> to real-world datasets</p> </li> <li> <p>Facilitate research in:</p> <ul> <li> <p>Computational social systems</p> </li> <li> <p>Urban computing</p> </li> <li> <p>Intelligent transportation</p> </li> <li> <p>Resource allocation in 5G/6G systems</p> </li> </ul> </li> </ul> <h2> Dataset Structure</h2> <p>The dataset consists of the following components:</p> <h3>1. <code>population.csv</code></h3> <ul> <li> <p>Synthetic individuals with demographic attributes:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>age</code></p> </li> <li> <p><code>gender</code></p> </li> <li> <p><code>occupation</code></p> </li> <li> <p><code>income_level</code></p> </li> </ul> </li> </ul> <h3>2. <code>mobility.csv</code></h3> <ul> <li> <p>Simulated mobility patterns:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>timestamp</code></p> </li> <li> <p><code>location_id</code></p> </li> <li> <p><code>activity_type</code> (work, home, leisure, etc.)</p> </li> </ul> </li> </ul> <h3>3. <code>social_network.csv</code></h3> <ul> <li> <p>Social interaction graph:</p> <ul> <li> <p><code>source_agent</code></p> </li> <li> <p><code>target_agent</code></p> </li> <li> <p><code>interaction_weight</code></p> </li> </ul> </li> </ul> <h3>4. <code>resource_usage.csv</code></h3> <ul> <li> <p>Resource consumption patterns:</p> <ul> <li> <p><code>agent_id</code></p> </li> <li> <p><code>timestamp</code></p> </li> <li> <p><code>resource_type</code> (energy, bandwidth, transport)</p> </li> <li> <p><code>usage_value</code></p> </li> </ul> </li> </ul> <h3>5. <code>environment.json</code></h3> <ul> <li> <p>Smart city configuration:</p> <ul> <li> <p>Zones</p> </li> <li> <p>Infrastructure nodes</p> </li> <li> <p>Simulation parameters</p> </li> </ul> </li> </ul> <h2>⚙️ Data Generation Methodology</h2> <p>The dataset is generated using a hybrid approach combining:</p> <ul> <li> <p><strong>Agent-Based Modeling (ABM)</strong> for simulating individual behavior</p> </li> <li> <p><strong>Graph-based modeling</strong> for social interactions</p> </li> <li> <p><strong>Generative AI techniques</strong> (e.g., probabilistic sampling / deep generative models)</p> </li> <li> <p><strong>Rule-based constraints</strong> to ensure realistic urban dynamics</p> </li> </ul> <h2> Key Features</h2> <ul> <li> <p>✔ Multi-modal (demographic, mobility, social, resource data)</p> </li> <li> <p>✔ Scalable (configurable population size)</p> </li> <li> <p>✔ Privacy-preserving (no real personal data)</p> </li> <li> <p>✔ Socially-aware behavior modeling</p> </li> <li> <p>✔ Suitable for benchmarking and simulation</p> </li> </ul> <h2> Potential Use Cases</h2> <ul> <li> <p>Smart city resource optimization</p> </li> <li> <p>Traffic and mobility prediction</p> </li> <li> <p>Social network analysis</p> </li> <li> <p>AI model benchmarking</p> </li> <li> <p>Policy simulation and decision support</p> </li> <li> <p>6G-enabled intelligent systems</p> </li> </ul> <h2> Data Size & Format</h2> <ul> <li> <p>Format: CSV / JSON</p> </li> <li> <p>Records: Configurable (default: 10,000 agents)</p> </li> <li> <p>Time span: Simulated over configurable time intervals</p> </li> </ul> <h2> Ethical Considerations</h2> <p>This dataset is <strong>fully synthetic</strong> and does not contain any real personal or sensitive data. It is designed to comply with ethical AI and data privacy standards.</p> <h2> License</h2> <p>This dataset is released under the <strong>Creative Commons Attribution 4.0 International (CC BY 4.0)</strong> license.</p> <h2> Citation</h2> <p>If you use this dataset, please cite:</p> <blockquote> <p>Agal, S. (2026). <em>SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems</em>. (Dataset)</p> </blockquote> <h2> Related Work</h2> <p>This dataset is associated with the research manuscript:</p> <p><strong>“SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems”</strong></p> <h2> Contact</h2> <p><strong>Dr. Sanjay Agal</strong><br>Professor & Head, Department of Artificial Intelligence and Data Science<br>Parul Institute of Engineering and Technology, India</p> <p>For queries: sanjay.agal32685@paruluniversity.ac.in</p> <h2> Future Work</h2> <p>Future releases will include:</p> <ul> <li> <p>Larger-scale simulations</p> </li> <li> <p>Real-time streaming data</p> </li> <li> <p>Integration with IoT sensor models</p> </li> <li> <p>Enhanced behavioral realism using advanced generative models</p> </li> </ul> <h2> Acknowledgment</h2> <p>We acknowledge the support of academic and research environments that facilitated the development of this dataset.</p> |
| title | SynthCity |
| url | https://doi.org/10.5281/zenodo.19104742 |