Publication

Abstract : Weather data analysis examines large volumes of weather-related information to derive meaningful insights and patterns. It involves collecting, processing, and analyzing various meteorological data, to gain a deeper understanding of weather patterns, trends, and anomalies. This study integrates temperature data with emissions records to explore the impacts of human activities on climate trends in India. Hadoop MapReduce and Pig Latin were used for preprocessing, while machine learning models including linear regression, Random Forest, Support Vector Regression, and ARIMA, forecasted temperature changes and identified significant climate drivers. Statistical tests like Mann-Kendall were performed to validate the trends observed in the data, adding robustness to the findings. Key contributions include identifying major emissions sectors, improving prediction accuracy, and revealing correlations between emissions, socioeconomic factors, and temperature changes. The ARIMA model, evaluated against recent temperature data from 2021 to 2023, achieved a Mean Absolute Percentage Error (MAPE) of 0.231%, demonstrating its reliability. Additionally, strong correlations were found between temperature variations and emissions from CO 2, methane, and nitrous oxide. Industrial emissions from manufacturing, construction, and transportation, as well as agricultural activities, were found to play critical roles in climate change, highlighting the need for focused climate policies.