Implementation of Multiple Linear Regression Algorithm to Predict Air Temperature Based on Pollutant Levels in South Tangerang City

Global warming is a phenomenon that has widespread effects, particularly on environmental aspects. Globally, the impact of global warming is evident in the rising temperatures of the Earth. In April 2023, much of South Asia experienced a heatwave with temperatures exceeding 40°C. In Indonesia, the daily maximum temperature recorded reached 37.2°C in South Tangerang City. Global warming is caused by the increasing concentration of greenhouse gases in the Earth's atmosphere. This study proposes a model for predicting air temperature by considering the influence of pollutant levels and daily climate data in South Tangerang City. The prediction modeling in this study uses the Multiple Linear Regression algorithm with an 80% training data and 20% testing data split. Out of 8 trials, the sixth model is the best with a k value of 8 and features including RH_avg, RR, ss, ddd_car, ff_avg, no2, o3, and pm10. The evaluation results of the sixth model yielded an R² value of 0.72749, MAE of 0.55593, MSE of 0.50078, and MAPE of 1.99806%.