Copyright: © the author(s), publisher and licensee Technoscience Academy. This is an open-access article distributed under the terms of the Creative Commons Attribution Non-Commercial License, which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited International Journal of Scientific Research in Science, Engineering and Technology Print ISSN: 2395-1990 | Online ISSN : 2394-4099 (www.ijsrset.com) doi : https://doi.org/10.32628/IJSRSET2310288 519 Marine Weather Forecasting to Enhance Fisherman’s Safety Using Machine Learning M. Robinson Joel, G. Manikandan, M. Nivetha Department of Information Technology, Kings Engineering College, Chennai, India A R T I C L E I N F O A B S T R A C T Article History: Accepted: 01 April 2023 Published: 19 April 2023 Through the use of scientific knowledge and weather measurements, weather forecasting is a technique for predicting what the atmosphere will be like in a certain location. In other words, it's a method by which the characteristics of a meteorological state are determined in advance by factors such as temperature, wind, humidity, rainfall, and the quantity of the data set. In an effort to foretell weather conditions now and in the future, meteorologists use a process called weather forecasting. For everyday operations, accurate weather forecasts are required, and this has been one of the most difficult problems to solve globally since the data is multidimensional and nonlinear. According to the survey, supervised and unsupervised machine learning algorithms, artificial neural networks, naive bayes algorithms, and random forest algorithms are some of the different techniques and algorithms utilised for weather prediction in the field of data mining. Keywords: Datamining, Weather prediction, Weather forecasting, SVM, Navie Publication Issue Volume 10, Issue 2 March-April-2023 Page Number 519-526 I. INTRODUCTION Traditionally, weather predictions have been made using huge, complicated physics models that take into account a variety of atmospheric circumstances over a lengthy period of time. The weather system's disturbances frequently make these circumstances unstable, which leads the models to provide erroneous forecasts. In a big High-Performance Computing (HPC) environment, the models are often run on hundreds of nodes, which uses a lot of energy. In this project, we describe a method for weather forecasting that uses historical data from a number of weather stations to train basic machine learning models, which can quickly and accurately anticipate specific weather conditions for the near future. The models may be used in contexts with a lot fewer resources. The evaluation's findings demonstrate that the models' accuracy is sufficient to be employed in conjunction with the most cutting-edge methods currently available. Furthermore, we demonstrate that using data from numerous nearby weather stations is preferable than using data from only the area for which weather forecasting is being done.