Skip to main content
UCI Machine Learning Repository

Gas Sensor Array Drift Dataset

Environment & Air Quality Environmental Sensing
1,728 views
2 min read
License
Catalog metadata: This page is a discovery record, not publisher documentation. Verify the description, schema, provenance, version, licence, and citation at the linked source before use.

Catalog Summary

"Long-term laboratory recordings from a 16-sensor chemical gas array exposed to six different gases across multiple batches, designed to study sensor drift and robustness of gas classification models.[web:87][web:89][web:95][web:104]"

Catalog Notes

Overview

The Gas Sensor Array Drift Dataset contains real measurements from an array of 16 metal-oxide gas sensors exposed to six distinct pure gases at different concentration levels over time.[web:87][web:89][web:95][web:104]

Data Collection

  • 13,910 measurements collected in ten batches over a period of 36 months, capturing long-term sensor drift behavior.[web:87][web:89][web:95]
  • Six gases: Ammonia, Acetaldehyde, Acetone, Ethylene, Ethanol, and Toluene.[web:87][web:95][web:104]
  • Each measurement is a vector of sensor responses plus a label indicating the gas class (and concentration for the related “different concentrations” variant).[web:87][web:89]

Features

  • 16 real-valued features corresponding to conductance-based responses of the metal-oxide sensors.[web:87][web:89]
  • Class labels for the six gases, enabling supervised classification and drift-compensation experiments.[web:87][web:95]
  • Batch identifiers to study temporal drift and domain adaptation across acquisition sessions.[web:87][web:89][web:104]

Use Cases

  • Benchmarking drift compensation and domain adaptation methods for electronic nose systems.[web:95][web:98]
  • Developing robust gas classification models for environmental monitoring and industrial safety.[web:87][web:95]
  • Analyzing long-term stability of IoT chemical sensor deployments.[web:87][web:89]

License and Terms

The dataset is distributed through the UCI Machine Learning Repository; users should follow the repository’s terms of use and any conditions specified on the dataset page.[web:89][web:70][web:65]

View Data Structure

To explore column names, data types, and sample rows, visit the official dataset page on UCI Machine Learning Repository.

Preview on UCI Machine Learning Repository

Cite This Dataset 

Vergara, A. E. A. (2012). Gas Sensor Array Drift Dataset. [Dataset]. UCI Machine Learning Repository. https://archive.ics.uci.edu/dataset/224/gas+sensor+array+drift+dataset

Source metadata: UCI Machine Learning Repository (2012)

Indexed by IoTDataset.com on Jan 27, 2026

Review the Source Record

Confirm the licence, version, access conditions, file format, and provenance at the source before use.

Open Source Page

Related Topics & Keywords

#UCI #gas sensors #electronic nose #sensor drift #environmental monitoring #IoT
Browse all Environment & Air Quality datasets

Share This Research

More in Environment & Air Quality

View All
Environmental Monitoring UCI Machine Learning Repository

UCI Air Quality Roadside Multisensor Dataset

Real-world hourly air quality measurements from an array of chemical gas sensors deployed at road level in a polluted Italian city, collected from March 2004 to February 2005.[web:46][web:53][web:77]

Jan 26, 2026
Environmental Monitoring / Air Quality IoT UCI Machine Learning Repository

Beijing Multi-Site Air-Quality Data

Hourly air pollutant measurements (PM2.5, PM10, SO₂, NO₂, CO, O₃) and meteorological data from 12 nationally controlled monitoring sites in Beijing, collected from March 2013 to February 2017 (420,768 instances).[page:1][web:74]

Jan 29, 2026
Environmental Kaggle

Gaoyou Lake IoT Pollution Risk: Water Quality Sensor Measurements [Environmental]

IoT environmental sensor dataset tracking water quality and pollution risk indicators for Gaoyou Lake, China. Includes pH, dissolved oxygen, turbidity, temperature, and conductivity readings. CSV format via Kaggle. Used for aquatic pollution risk classification and IoT water monitoring research.

May 03, 2026
Environmental UCI

UCI Air Quality: Italian City Chemical Multisensor Device [9,358 Hourly Records]

Longest freely available on-field IoT air quality sensor deployment: 9,358 hourly records from 5 metal oxide gas sensors in an Italian city. CSV/XLSX format. Used for gas sensor regression, drift correction, and pollution forecasting research.

Apr 17, 2026
Environmental Kaggle

Integrated Satellite-IoT-Machine Learning Framework for Disaster Management

Published in November 2025, this dataset provides a multimodal framework integrating satellite imagery and IoT sensor data for environmental monitoring and disaster management. It is designed to support the development of machine learning models that synchronize remote sensing with ground-based IoT observations for real-time risk assessment.

Feb 06, 2026
Environmental Kaggle

AIR4LIFE: Dual-Node Environmental Monitoring Dataset

AIR4LIFE is a high-resolution air quality monitoring dataset published in late 2025, featuring measurements from a dual-node IoT setup. It tracks pollutants and ambient conditions to evaluate the trade-offs between energy-aware duty cycling and data completeness in environmental sensing networks.

Feb 06, 2026

Explore other topics

All topics →
Healthcare & Medical IoT 31 Smart Agriculture 21 Smart City & Urban Sensing 59 Smart Home & Buildings 79 Industrial IoT & Predictive Maintenance 48 Energy & Smart Grid 14 Wearables & Human Activity 7 Connected Vehicles 7 IoT Security & Intrusion Detection 57 Sensor Networks & Telemetry 14