LSTM model to forecast outdoor air pollutants
Patrick Duhirwe
Data Analyst
ML Engineer
Python
R
TensorFlow
Data Preprocessing: Cleaning and preparing the data for analysis, including normalization and handling missing values.
LSTM Model Development: Developing Long Short-Term Memory (LSTM) models to forecast urban air pollutants based on historical meteorology data.
Performance Testing: Testing the performance of 110 different LSTM models across various pollutants and conditions.
Analysis of Sensor Requirements: Investigating the necessary number of meteorological sensors for effective air quality management.
Impact Assessment of Extreme Conditions: Assessing how extreme conditions such as bushfires and COVID-19 lockdowns affect pollutant predictability.
Forecasting Capability Exploration: Exploring the forecasting capabilities of models under standard and non-standard conditions.
