Articles | Volume 11, issue 1
Atmos. Meas. Tech., 11, 291–313, 2018
https://doi.org/10.5194/amt-11-291-2018
Atmos. Meas. Tech., 11, 291–313, 2018
https://doi.org/10.5194/amt-11-291-2018

Research article 15 Jan 2018

Research article | 15 Jan 2018

A machine learning calibration model using random forests to improve sensor performance for lower-cost air quality monitoring

Naomi Zimmerman et al.

Download

Interactive discussion

Status: closed
Status: closed
AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment
Printer-friendly Version - Printer-friendly version Supplement - Supplement

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision
AR by Naomi Zimmerman on behalf of the Authors (03 Nov 2017)  Author's response    Manuscript
ED: Publish subject to technical corrections (20 Nov 2017) by Dominik Brunner
Download
Short summary
Low-cost sensors promise neighborhood-scale air quality monitoring but have been plagued by inconsistent performance for precision, accuracy, and drift. CMU and SenSevere collaborated to develop the RAMP, which uses electrochemical sensors. We present a machine learning algorithm that overcomes previous performance issues and meets US EPA's data quality recommendations for personal exposure for NO2 and tougher "supplemental monitoring" standards for CO & ozone across 19 RAMPs for several months.