SynthCity

SynthCity Dataset A Multi-Modal Synthetic Dataset for Socially Realistic Smart City Systems 📌 Overview The SynthCity dataset is a synthetic, multi-modal dataset generated using the SynthCity framework, designed to simulate realistic human behavior, mobility patterns, and social interactions in AI-driven smart city environments. 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. 🎯 Purpose Support reproducible research in smart city analytics Enable benchmarking of AI/ML models under realistic social conditions Provide privacy-preserving alternative to real-world datasets Facilitate research in: Computational social systems Urban computing Intelligent transportation Resource allocation in 5G/6G systems 📂 Dataset Structure The dataset consists of the following components: 1. population.csv Synthetic individuals with demographic attributes: agent_id age gender occupation income_level 2. mobility.csv Simulated mobility patterns: agent_id timestamp location_id activity_type (work, home, leisure, etc.) 3. social_network.csv Social interaction graph: source_agent target_agent interaction_weight 4. resource_usage.csv Resource consumption patterns: agent_id timestamp resource_type (energy, bandwidth, transport) usage_value 5. environment.json Smart city configuration: Zones Infrastructure nodes Simulation parameters ⚙️ Data Generation Methodology The dataset is generated using a hybrid approach combining: Agent-Based Modeling (ABM) for simulating individual behavior Graph-based modeling for social interactions Generative AI techniques (e.g., probabilistic sampling / deep generative models) Rule-based constraints to ensure realistic urban dynamics 📊 Key Features ✔ Multi-modal (demographic, mobility, social, resource data) ✔ Scalable (configurable population size) ✔ Privacy-preserving (no real personal data) ✔ Socially-aware behavior modeling ✔ Suitable for benchmarking and simulation 🔬 Potential Use Cases Smart city resource optimization Traffic and mobility prediction Social network analysis AI model benchmarking Policy simulation and decision support 6G-enabled intelligent systems 📏 Data Size & Format Format: CSV / JSON Records: Configurable (default: 10,000 agents) Time span: Simulated over configurable time intervals 🔐 Ethical Considerations This dataset is fully synthetic and does not contain any real personal or sensitive data. It is designed to comply with ethical AI and data privacy standards. 📜 License This dataset is released under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. 📖 Citation If you use this dataset, please cite: Agal, S. (2026). SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems. (Dataset) 🔗 Related Work This dataset is associated with the research manuscript: “SynthCity: A Multi-Modal Generative Framework for Socially Realistic Synthetic Data in AI-Driven Smart City Systems” 📬 Contact Dr. Sanjay AgalProfessor & Head, Department of Artificial Intelligence and Data ScienceParul Institute of Engineering and Technology, India For queries: [e-mail] 🚀 Future Work Future releases will include: Larger-scale simulations Real-time streaming data Integration with IoT sensor models Enhanced behavioral realism using advanced generative models 🙏 Acknowledgment We acknowledge the support of academic and research environments that facilitated the development of this dataset.