MACHINE LEARNING FOR COST ESTIMATION AND FORECASTING IN BANKING: A COMPARATIVE ANALYSIS OF ALGORITHMS

Deleted Journal, Год журнала: 2024, Номер 04(12), С. 6 - 17

Опубликована: Дек. 9, 2024

Accurate cost estimation and forecasting are critical for effective decision-making in the banking sector. This study evaluates performance of machine learning algorithms, including Linear Regression, Ridge Random Forest, Gradient Boosting Machine (GBM), Long Short-Term Memory (LSTM) networks, prediction using a robust dataset comprising operational, transactional, macroeconomic features. Our results demonstrate that while simpler models like Regression offer computational efficiency, their predictive accuracy is limited handling complex data. Tree-based methods, particularly Forest GBM, significantly enhance by capturing intricate patterns, albeit at higher cost. The LSTM network outperformed all models, achieving highest R² score lowest MAE MSE values, highlighting its superiority temporal dependencies. research provides actionable insights institutions, emphasizing trade-offs between accuracy, model complexity. findings pave way optimized ML adoption financial forecasting, enhancing operational resilience strategic planning.

Язык: Английский

A Comparative Study of Brain Tumor Detection using Convolutional Neural Networks with MRI Images

Опубликована: Ноя. 29, 2023

This research emphasizes the global health challenge of brain tumors and importance early detection using Convolutional Neural Networks (CNNs) on Magnetic Resonance Imaging (MRI). The dataset, including healthy tumor MRI scans, underwent careful processing for CNN input. With a SoftMax Fully Connected layer, achieved 98% accuracy, outperforming Radial Basis Function (RBF) Decision Tree (DT) classifiers. Feature extraction through clustering improved with classifier reaching 99.52% test data. study advances deep learning in medical image analysis, highlighting CNN-MRI synergy precise potential advancements treatment patient care.

Язык: Английский