Home / Current Issue / Paper 1705479
Climatic Temperature Trends: Analyzing Past Data and Predicting Future Temperature
Subject area: Science,Engineering and Technology · Area of research: Information Technology
Abstract
Machine Learning (ML) techniques for time series prediction are becoming increasingly accurate and helpful, particularly in considering climate change, such as weather forecasting, climate research, and environmental planning, accurate temperature prediction is essential. This work utilizes a 40-year historical dataset and three machine learning algorithms, Linear Regression (LR), Random Forest (RF) and Polynomial Regression to give a thorough method to temperature prediction. To increase the accuracy of temperature forecasts, it is important to examine how well various algorithms work on a large dataset. A large historical dataset covering four decades is first gathered and pre-processed for the study. This dataset includes temperature records as well as data on a wide range of meteorological factors, including humidity, wind speed, and precipitation. The dataset is optimized for modeling using feature selection and engineering methods. The first step in predicting temperature is to use linear regression, polynomial Regression - Single modeling approach and use random forest algorithm ? Ensemble modeling and for the same and then compare these algorithms. The data has been collected via NASA climate datasets. It consists of 40 years data from 1980 to 2020 in month wise format.
Keywords
Machine learning, Temperature prediction, Ensemble Models, Single Models.
How to cite this paper
@article{1705479,
author = {Rishika Rao, Harsh Panchal, Dr. S. K. Singh},
title = {Climatic Temperature Trends: Analyzing Past Data and Predicting Future Temperature},
journal = {Iconic Research And Engineering Journals},
year = {2024},
volume = {7},
number = {8},
pages = {166-170},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1705479.pdf},
abstract = {Machine Learning (ML) techniques for time series prediction are becoming increasingly accurate and helpful, particularly in considering climate change, such as weather forecasting, climate research, and environmental planning, accurate temperature prediction is essential. This work utilizes a 40-year historical dataset and three machine learning algorithms, Linear Regression (LR), Random Forest (RF) and Polynomial Regression to give a thorough method to temperature prediction. To increase the accuracy of temperature forecasts, it is important to examine how well various algorithms work on a large dataset. A large historical dataset covering four decades is first gathered and pre-processed for the study. This dataset includes temperature records as well as data on a wide range of meteorological factors, including humidity, wind speed, and precipitation. The dataset is optimized for modeling using feature selection and engineering methods. The first step in predicting temperature is to use linear regression, polynomial Regression - Single modeling approach and use random forest algorithm ? Ensemble modeling and for the same and then compare these algorithms. The data has been collected via NASA climate datasets. It consists of 40 years data from 1980 to 2020 in month wise format.},
keywords = {Machine learning, Temperature prediction, Ensemble Models, Single Models.},
month = {February},
}