Comparison of Average Temperature Prediction Based on XGBoost and Prophet Models

Xiang Zhou

doi:10.54097/x2s0s641

Authors

Xiang Zhou

DOI:

https://doi.org/10.54097/x2s0s641

Keywords:

Temperature prediction; XGBoost; Prophet.

Abstract

Temperature prediction is extremely important in various fields. With this in mind, this study utilized two prominent models, XGBoost and Prophet, known for their efficacy in time series forecasting, to perform daily average temperature predictions based on historical data. XGBoost model enhances the Gradient Boosted Decision Tree method and the Prophet model leverages an additive model that combines seasonal and holiday features. The dataset comprised historical data (1995-2020) of daily average temperatures from major cities, serving as an ideal time series subject for forecasting. This study trained both models on the initial 80% of the data from four cities and used them to forecast the remaining period. This research employed Root Mean Square Error (RMSE) analysis to evaluate the performance of the models. The results indicated that the Prophet model surpassed the XGBoost model in two out of the four cities and matched its performance in the other two.

Downloads

Download data is not yet available.

References

Lewis F R. Weather prediction by numerical process. Hardpress Publishing, 2012.

Bcochenk B, Ustrnul Z. Machine Learning in Weather Prediction and Climate Analyses—Applications and Perspectives. Atmosphere, 2022, 13(2): 180.

Cohen J, Coumou D, Hwang J, et al. S2S reboot: An argument for greater inclusion of machine learning in subseasonal to seasonal forecasts Wiley Interdisciplinary Reviews: Climate Change, 2018, 10(2): 10-15.

Kramer F A, Palma D, Braten A E, etc. Operationalizing Solar Energy Predictions for Sustainable, Autonomous IoT Device Management. IEEE Internet of Things Journal, 2021, 1: 1.

Potocnik P, Vidrih B, Kitanovski A, et al. Neural network, ARX, and extreme learning machine models for the short-term prediction of temperature in buildings. Building Simulation, 2019, 12(6): 1077-1093.

Rahman A, Abbas S, Gollapalli M, et al. Rainfall Prediction System Using Machine Learning Fusion for Smart Cities. Sensors, 2022, 22(9): 3504.

Barque M, Martin S, Vianin N, et al. Improving wind power prediction with retraining machine learning algorithms.Jumps, 2018

Taylor S J, Letham B. Forecasting at scale. 2017.

Daily Temperature of Major Cities. Retrieved from: https://www.kaggle.com/datasets/sudalairajkumar/daily-temperature-of-major-cities.

Chen T, Guestrin C. Xgboost: A scalable tree boosting system. Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining. 2016: 785-794.