UrbanIoT-Anomaly: Multimodal Smart City Dataset for Urban Safety (2025)
Abstract
"Multimodal smart city dataset combining IoT sensor data (environmental, traffic, infrastructure) with ground-truth anomaly labels for urban safety applications, anomaly detection, and multi-source data fusion research in smart cities."
Description
Overview
The UrbanIoT-Anomaly dataset is a multimodal smart city dataset designed for anomaly detection and urban safety research, integrating diverse IoT sensor streams with labeled anomalous events.
Data Sources and Modalities
- Environmental sensors: Air quality (PM2.5, PM10, CO, NO2), temperature, humidity, and noise levels from distributed monitoring stations.
- Traffic sensors: Vehicle flow, speed, and occupancy data from road-embedded sensors and cameras.
- Infrastructure sensors: Smart streetlight status, parking occupancy, waste bin fill levels, and utility consumption.
- Event logs: Timestamped records of incidents, maintenance activities, and public safety events.
Anomaly Types
- Environmental anomalies: Air quality spikes, abnormal temperature/humidity patterns, excessive noise events.
- Traffic anomalies: Congestion, accidents, unusual flow patterns, and parking violations.
- Infrastructure anomalies: Sensor malfunctions, streetlight outages, and abnormal utility consumption.
- Multi-source anomalies: Correlated events across multiple sensor types requiring fusion for detection.
Dataset Characteristics
- Real-world data collected from a large urban area over an extended monitoring period.
- Ground-truth anomaly labels derived from official incident reports, expert annotations, and automated validation.
- Temporal resolution ranging from minutes (traffic, environmental) to hourly (infrastructure) depending on sensor type.
- Spatial coverage across multiple urban zones enabling spatial correlation analysis.
Use Cases
- Anomaly detection: Developing and benchmarking multimodal anomaly detection algorithms for smart city applications.
- Urban safety: Early warning systems for environmental hazards, traffic incidents, and infrastructure failures.
- Data fusion: Research on integrating heterogeneous IoT sensor streams for improved situational awareness.
- Smart city analytics: Understanding urban dynamics and cross-domain correlations in city operations.
View Data Structure
To explore column names, data types, and sample rows, visit the official dataset page on Kaggle / Research Square Preprint.
Preview on Kaggle / Research Square PreprintCite This Dataset
Ziya, et al. (2025). UrbanIoT-Anomaly: Multimodal Smart City Dataset for Urban Safety (2025). [Dataset]. Kaggle. https://www.kaggle.com/datasets/ziya07/urbaniot-anomaly-multimodal-smart-city-dataset
Source: Kaggle (2025)
Indexed by IoTDataset.com on Jan 30, 2026
Ready to Start Your Research?
Download this dataset directly from the official repository and start building your next breakthrough project.