0
- Volume 1, Issue 1 2023
By Hira Farman, Noman Islam
10.20547/aibd.231102
Keywords: Precipitation, Weather forecast, Machine Learning, DL.
Deep neural networks are now widely employed in artificial intelligence applications that have significantly changed human livelihoods in a number of ways. Weather forecasting has long been considered an important area of study with far-reaching consequences for disaster management and public safety. This research analyzes the use of deep learning and machine learning techniques to create a mobile application that offers users real-time weather predictions. The primary goal of this study is to look into the usefulness of machine learning and deep learning algorithms in predicting weather patterns using historical weather data Rainfall. To analyze various data sources and develop weather forecasts, Artificial Neural Networks (ANN), Random forest, K-Means clustering and linear regression techniques are used. To illustrate the feasibility of Machine learning and deep learning models in weather prediction, this research focuses on model construction, training, and evaluation. The Weather Underground dataset for Pakistan was gathered. In addition to using deep learning and machine learning techniques, Weather Underground is a commercial weather service that offers real-time meteorological information. He foundation of this research work is a proprietary dataset that was carefully collected from Wunderground.com and has over 71,000 records spanning five years. The results of this study showed that artificial neural networks (ANNs)performed at the greatest levels, instead of k means clustering, random forest classification, and linear regression. The accuracy values for random forest, linear regression, 0.76, and ANN has 0.85, respectively. The significance of data quality, model correctness, data gathering, and societal benefits is emphasized by this study. It lays the groundwork for future development, which could lead to the establishment of a comprehensive mobile application that combines precise weather forecasts with different catastrophe awareness elements.