Finding Weather Prediction Accuracy Using Linear Regression

Abstract

Weather forecasting is a technology to found the future condition of the atmosphere for the selected location. This research field is the most sensitive for real-time issues. We propose a linear regression algorithm to predict “Rainfall”, “Max Temp”, “Min Temp” and “Relative Humidity”. We are using here Jupyter Notebook and library was Numpy, Pandas for finding prediction accuracy. We find this accuracy from Bangladesh weather data perspective and finding annual accuracy of weather prediction using linear regression from 1949-2013.proper weather forecasting is very important for our daily life and our country. It affects our daily lives and our economy. The weather prediction that we have can be used to accurately predict the weather. To deliver better weather forecasting, we have to learn machine learning techniques perfectly. Machine learning techniques will help us to analyze weather prediction patterns from the dataset. In this research paper, we can use a machine learning-based weather prediction method and lead dataset to analyze real-time humidity, maximum temperature, minimum temperature and rainfall. In this paper, we use a direct linear regression method t calculate the accuracy of rainfall, maximum temperature, minimum temperature and humidity. Based on this method we calculate not only the relation between dataset and prediction accuracy of weather forecasting but also the relation between dataset modernism and prediction accuracy.