Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Precision of Automated Teller Machines Network Quality Assessment: Machine Learning and Multi Classifier Fusion Approaches

Alireza Safarzadeh, Mohammad Reza Jamali, Behzad Moshiri

TL;DR

This paper addresses the problem of unreliable ATM status detection and false alarms by formulating it as a binary classification task and proposing a multi-classifier fusion framework centered on a Stacking Classifier. By balancing data with SMOTE and integrating diverse models (Random Forest, LightGBM, CatBoost), the approach achieves a dramatic reduction in false alarms (from 3.56% to 0.71%) and an overall accuracy of 99.29%. Extensive experiments compare SVM, Decision Tree, Bagging, RF, LightGBM, CatBoost, DCS, DES, and Stacking, with the stacking ensemble delivering the best performance. The results demonstrate the practical value of data fusion and SMOTE in enhancing ATM network reliability, reducing maintenance costs, and improving customer satisfaction, while outlining avenues for real-time monitoring and cross-domain applications.

Abstract

Ensuring reliable ATM services is essential for modern banking, directly impacting customer satisfaction and the operational efficiency of financial institutions. This study introduces a data fusion approach that utilizes multi-classifier fusion techniques, with a special focus on the Stacking Classifier, to enhance the reliability of ATM networks. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied, enabling balanced learning for both frequent and rare events. The proposed framework integrates diverse classification models - Random Forest, LightGBM, and CatBoost - within a Stacking Classifier, achieving a dramatic reduction in false alarms from 3.56 percent to just 0.71 percent, along with an outstanding overall accuracy of 99.29 percent. This multi-classifier fusion method synthesizes the strengths of individual models, leading to significant cost savings and improved operational decision-making. By demonstrating the power of machine learning and data fusion in optimizing ATM status detection, this research provides practical and scalable solutions for financial institutions aiming to enhance their ATM network performance and customer satisfaction.

Enhancing Precision of Automated Teller Machines Network Quality Assessment: Machine Learning and Multi Classifier Fusion Approaches

TL;DR

This paper addresses the problem of unreliable ATM status detection and false alarms by formulating it as a binary classification task and proposing a multi-classifier fusion framework centered on a Stacking Classifier. By balancing data with SMOTE and integrating diverse models (Random Forest, LightGBM, CatBoost), the approach achieves a dramatic reduction in false alarms (from 3.56% to 0.71%) and an overall accuracy of 99.29%. Extensive experiments compare SVM, Decision Tree, Bagging, RF, LightGBM, CatBoost, DCS, DES, and Stacking, with the stacking ensemble delivering the best performance. The results demonstrate the practical value of data fusion and SMOTE in enhancing ATM network reliability, reducing maintenance costs, and improving customer satisfaction, while outlining avenues for real-time monitoring and cross-domain applications.

Abstract

Ensuring reliable ATM services is essential for modern banking, directly impacting customer satisfaction and the operational efficiency of financial institutions. This study introduces a data fusion approach that utilizes multi-classifier fusion techniques, with a special focus on the Stacking Classifier, to enhance the reliability of ATM networks. To address class imbalance, the Synthetic Minority Over-sampling Technique (SMOTE) was applied, enabling balanced learning for both frequent and rare events. The proposed framework integrates diverse classification models - Random Forest, LightGBM, and CatBoost - within a Stacking Classifier, achieving a dramatic reduction in false alarms from 3.56 percent to just 0.71 percent, along with an outstanding overall accuracy of 99.29 percent. This multi-classifier fusion method synthesizes the strengths of individual models, leading to significant cost savings and improved operational decision-making. By demonstrating the power of machine learning and data fusion in optimizing ATM status detection, this research provides practical and scalable solutions for financial institutions aiming to enhance their ATM network performance and customer satisfaction.
Paper Structure (28 sections, 2 equations, 8 figures, 13 tables)

This paper contains 28 sections, 2 equations, 8 figures, 13 tables.

Table of Contents

  1. Introduction
  2. Literature Review
  3. Methodology
  4. Problem Formulation
  5. Feature Extraction
  6. Kolmogorov-Smirnov Test and Probability Threshold
  7. Probability Threshold Justification
  8. Data Imbalance and SMOTE
  9. Machine Learning Models
  10. Evaluation Metrics
  11. Experimental Results
  12. Feature Extraction and Correlation
  13. ATM Status Files and Extracted Transaction Status Performance
  14. Effect of SMOTE on Classification Performance
  15. Classification Models and Performance
  16. ...and 13 more sections

Figures (8)

  • Figure 1: Schematic representation of the ATM network within the banking system, illustrating data flows between ATMs, status files, transaction records, and the central monitoring system.
  • Figure 2: Example of false alarm and missed alarm scenarios.
  • Figure 3: Extraction of out-of-service status from log journal based on errors in card reader, keypad, and communication disconnection.
  • Figure 4: Kolmogorov-Smirnov test results for transaction intervals.
  • Figure 5: Occurrence of the transaction at moment zero with an average distance between two transactions of 271 seconds (Left) vs. Extracted Status (Right).
  • ...and 3 more figures