Main goal of DeepUrban in KAIST
We specialize in providing meticulously curated and data-engineered datasets that are specifically designed for AI research. Our datasets are carefully formatted and organized to ensure ease of use and compatibility with various AI applications.
In addition to our high-quality datasets, we also offer a comprehensive collection of sample codes that serve as invaluable resources, demonstrating how to effectively utilize our data for AI research purposes. These code examples provide valuable insights and practical guidance, empowering researchers to explore and leverage our datasets to their full potential.
Our goal is to simplify the data acquisition and utilization process for AI researchers, enabling them to focus on their core research objectives without being burdened by data engineering complexities. With our data and accompanying sample codes, researchers can seamlessly integrate our datasets into their AI projects, accelerating their research progress and fostering groundbreaking discoveries in the field of artificial intelligence.
Research topic
The KAIST DeepUrban Group specializes in computational urban science, focusing on four key areas related to cities: monitoring, prediction, simulation, and user interaction.
Monitoring
- Utilizing Geographic Information Systems (GIS) and database systems (DB) to collect and analyze urban data.
- Developing techniques for urban change detection to identify and understand changes in urban landscapes, infrastructure, and socio-economic indicators.
Prediction
- Applying machine learning and deep learning algorithms to predict various urban phenomena.
- Training models on historical data to forecast outcomes such as population growth, traffic congestion, energy consumption, and air quality.
Simulation
- Leveraging data science methodologies to simulate complex urban systems and discover patterns within them.
- Integrating diverse datasets to create computational models that simulate urban dynamics and enable scenario analysis and policy evaluation.
User Interaction
- Enhancing human-computer interaction (HCI) in the urban context through interactive interfaces, visualization tools, and user-centered applications.
- Promoting citizen involvement for urban decision making, allowing citizens, researchers, and policymakers to actively participate in shaping the future of cities.