Research map / Large language models / AI and society
Indoor Air Quality Dataset with Activities of Daily Living in Low to Middle-income Communities
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Releases DALTON, a six-month indoor air quality dataset from 30 sites across rural, suburban and urban India, with pollutant sensor streams and speech-to-text activity annotations.
Problem
Indoor air pollution in developing countries like India is under-studied, and existing indoor air quality datasets are small, scenario-specific, or focused on comfort and energy rather than pollution.
Approach
A low-cost multi-sensor platform (DALTON) deployed across 30 indoor sites in four Indian regions, paired with a speech-to-text Android app (vocalAnnot) that lets occupants annotate activities in real time aligned with pollution change-points.
Key results
- Dataset covers 30 indoor sites across four Indian regions over six months (summer and winter), totaling about 89.1M samples and 13,646 hours of air quality data.
- Contains 3,957 activity annotations from 24 of 46 occupants, collected via the vocalAnnot speech-to-text app.
- Measures eight pollutants plus temperature and humidity at 1 Hz using the DALTON sensor module.
Surprising findings
- Occupants appear to ignore ventilation when temperature is comfortable, even as pollutants accumulate to harmful levels (hypothesis from data).
Limitations
- Many annotations are missing or delayed, so some activity labels are unreliable.
- Frequent power outages cause gaps; readings are interpolated only for outages under 15 minutes.
- Only six months of data; monsoon and autumn are not captured and studio apartments limit spread measurement.
Experimental setup
Datasets: DALTON indoor air quality dataset
Builds on: KLCPD ruptures
Closest papers by meaning
- AirRadar: Inferring Nationwide Air Quality in China with Deep Neural Networks (AAAI 2025 · similarity 0.78)
- How Your Location Relates to Health: Variable Importance and Interpretable Machine Learning for Environmental and Sociodemographic Data (AAAI 2025 · similarity 0.77)
- UrbanDataLayer: A Unified Data Pipeline for Urban Science (NeurIPS 2024 · similarity 0.76)
- Measuring Fine-Grained Urban Air Temperature with Satellite Imagery (AAAI 2025 · similarity 0.76)
- Building Timeseries Dataset: Empowering Large-Scale Building Analytics (NeurIPS 2024 · similarity 0.76)
- Terra: A Multimodal Spatio-Temporal Dataset Spanning the Earth (NeurIPS 2024 · similarity 0.75)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.