Comparative Evaluation of Weather Forecasting using Machine Learning Models
Md Saydur Rahman, Farhana Akter Tumpa, Md Shazid Islam, Abul Al Arabi, Md Sanzid Bin Hossain, Md Saad Ul Haque
TL;DR
This study assesses multiple machine learning approaches for long-term weather forecasting using a 20-year Dhaka dataset. By evaluating six models on two targets—precipitation and average temperature—the work provides detailed performance comparisons, with AdaBoost frequently delivering the best accuracy and balanced precision/recall. The analysis highlights significant correlations between weather variables and seasonal patterns, and concludes that ensemble and boosting methods hold substantial promise for improving forecast reliability. The authors also outline concrete directions for future work, including feature selection, data augmentation, and integration of diverse data sources to bolster predictive capabilities in real-world settings.
Abstract
Gaining a deeper understanding of weather and being able to predict its future conduct have always been considered important endeavors for the growth of our society. This research paper explores the advancements in understanding and predicting nature's behavior, particularly in the context of weather forecasting, through the application of machine learning algorithms. By leveraging the power of machine learning, data mining, and data analysis techniques, significant progress has been made in this field. This study focuses on analyzing the contributions of various machine learning algorithms in predicting precipitation and temperature patterns using a 20-year dataset from a single weather station in Dhaka city. Algorithms such as Gradient Boosting, AdaBoosting, Artificial Neural Network, Stacking Random Forest, Stacking Neural Network, and Stacking KNN are evaluated and compared based on their performance metrics, including Confusion matrix measurements. The findings highlight remarkable achievements and provide valuable insights into their performances and features correlation.