Rainfall Classification Model for the Philippines using Optimized K-nearest Neighbor Algorithm with GridSearchCV Hyperparameter Tuning

How much rain is likely to fall depends on a number of important things, such as the temperature, the number of clouds, the humidity, the speed of the wind, and the direction of the wind. Because it’s hard to tell how much rain will fall, this study paper was written. It’s important to know when it will rain, especially in a farming area like the Philippines, where it rains at strange times. The purpose of this research was to develop a method that could be used in place of the Philippine Atmospheric, Geophysical, and Astronomical Services Administration’s daily weather forecasts to determine whether or not it will rain in the Philippines. In this analysis, the K-Nearest Neighbor machine learning technique is used to categorize the input data into four classes, corresponding to the absence of rain, the presence of light rain, moderate rain, and heavy rain. Success may be evaluated in a number of ways, including via the use of metrics like accuracy, memory, f1-score, precision, and the confusion matrix. Lastly, their classification performance is evaluated and improved by fine-tuning the settings of K-NN using grid search, which gives a better mean accuracy of 85%.