The development of Machine Learning methods to correct data responses from low-cost sensors to improve agricultural productivity and air quality data accuracy.

该项目旨在开发基于机器学习的方法，使用于交通、健康和农业的空气质量监测中的电化学和固态气体传感器的响应标准化。所提出的模型使用广泛的环境参数和参考级痕量气体分析仪以及电化学和固态传感器的响应作为训练集。预期的结果是低成本的气体传感器将报告具有较小不确定性的数据。因此，以更低的成本获得更好的数据。

This project is to develop machine learning based methods to normalise the responses of electrochemical and solid state gas sensors used in air quality monitoring for transport, health and agriculture.The proposed models use a wide range of environmental parameters and reference grade trace gas analysers as well as the responses of the electrochemical and solid state sensors as the training sets.The intended result is low cost gas sensors that will report data that with smaller uncertainties. Therefore better data at lower cost.