Ai for anomaly detection in glacier movement identifying climate change effect using machine learning

This paper discusses the use of artificial intelligence for anomaly detection in glacier movement to determine the effect of climate change. Using machine learning algorithms such as Logistic Regression, KNN, Random Forest, SVMs, and an Ensemble Model with XGBoost and LightGBM, the research seeks to improve the accuracy and reliability of anomaly detection. Data was obtained from Kaggle, with 15,000 records and 20 attributes. Extensive preprocessing techniques, such as missing data handling, feature engineering, and outlier detection, were utilized to enhance model performance. Results show that the Ensemble Model performs better than all other algorithms, with an impressive accuracy of 96%, in addition to high precision (94%), recall (95%), and F1 score (92%). These results indicate the efficacy of sophisticated ensemble methods in identifying anomalies, opening doors to more precise climate change impact assessments. This method provides significant information regarding environmental monitoring and risk assessment in glacier dynamics.