Deep-learning frameworks can effectively forecast the air pollution data for individual stations by decoding time-series data. However, most of the existing time-series-based deep-learning models use offline spatial interpolation strategies and thus cannot reliably project the station-based forecast to the spatial region of interest. In this project, based on our previously developed LSTM framework, the students are expected to take the factors which influence the air pollutants concentration (e.g., wind and temperature) into consideration and update our previously developed framework. Priority will be given to the students who are familiar with python and the LSTM deep-learning technique. Depending on the performance, the students will be included when the academic paper is published, which is highly important for the future graduate school application.