Abstract

Meteorological predictions play a pivotal role in various sectors, from agriculture to disaster management. While traditional weather prediction models exhibit proficiency, challenges persist in accurately capturing the complex and dynamic nature of atmospheric phenomena. Conventional weather prediction models often struggle to adapt to the intricacies of climate patterns, leading to suboptimal forecasting accuracy. The need for more robust methodologies that can effectively extract patterns from vast datasets and optimize model parameters is evident. Existing literature lacks comprehensive studies that seamlessly integrate ACO and Data Mining for weather prediction. This research bridges the gap by proposing a novel framework that leverages ACO optimization capabilities to refine Data Mining models, thereby improving the precision of weather forecasts. The proposed method involves utilizing ACO to optimize the parameters of Data Mining algorithms, such as decision trees and neural networks. ACO ability to find optimal solutions is harnessed to fine-tune the model parameters, enhancing its capability to extract meaningful patterns from historical weather data. Experiments demonstrate promising results, with a significant improvement in the accuracy of weather predictions compared to traditional models. The integrated approach shows particular efficacy in handling non-linear relationships and abrupt changes in weather patterns.