Journal of Computers and Digital Business

Banking Cybersecurity: Safeguarding Financial Information in the Digital Era

2025-05-01T13:08:57+07:00

This study explores the escalating cybersecurity challenges in the banking sector and the potential of large language models (LLMs) to enhance digital defense mechanisms. Employing a qualitative methodology that includes a systematic literature review, expert interviews, and case study evaluations, the research investigates the integration of LLMs in cybersecurity operations such as threat detection, automated incident response, and user authentication. The findings reveal that LLMs offer significant advantages in real-time anomaly detection, predictive analytics, and natural language-based security training. However, their adoption is hindered by concerns over algorithmic transparency, data privacy, and the need for specialized technical expertise within financial institutions. A key contribution of this work is the development of an integrated cybersecurity framework that combines AI-driven technologies, blockchain-based transaction security, digital forensic tools, and human-centered security practices. The proposed framework aims to guide financial institutions in implementing adaptive, intelligent cybersecurity strategies aligned with evolving global regulatory standards. This research offers both theoretical insights and practical recommendations for enhancing cyber resilience in digital banking environments. It emphasizes the importance of a multidimensional approach that addresses technical innovation, organizational preparedness, and regulatory compliance. Future studies are encouraged to validate the proposed framework through empirical testing across diverse banking infrastructures.

Machine Learning (ML) Algorithms for Diagnosing Blood Cancer in Blood Smear Images

2025-02-21T23:54:21+07:00

Artificial intelligence (AI), particularly deep learning (DL), has significantly advanced medical image analysis, including the detection and classification of blood cancer through blood smear images. This review explores the state-of-the-art data mining (DM) and DL techniques applied in the identification and classification of white blood cells (WBCs), with a focus on leukemia diagnosis. By systematically analyzing relevant literature from 2014 to 2024, the study highlights key AI algorithms, including traditional machine learning models such as SVM, KNN, and ANN, as well as modern DL architectures like CNN, RCNN, ResNet, and hybrid models. The review evaluates their performance, clinical applicability, and implementation challenges. Particular attention is given to the strengths of DL in feature extraction and classification accuracy, which often surpass traditional DM approaches. Despite these advances, issues such as data scarcity, computational cost, and the need for medical expertise remain major challenges. The study also outlines future directions involving lightweight DL models, transfer learning, and open-access datasets to enhance clinical deployment. Ultimately, this work provides a comprehensive foundation for researchers and developers aiming to improve blood cancer diagnosis through automated medical imaging systems powered by AI.

Network Log Implementation for GRU Based Bandwidth Classification

2025-04-11T09:31:12+07:00

Network bandwidth management using log data is a challenging task, especially in anomaly detection, e.g., fraudulent bandwidth that violates the Service Level Agreement (SLA). The present study suggests a deep learning automatic classification method for network logs, which leverages the Gated Recurrent Unit (GRU) and is used in time-series tensor configurations given as [N, 5, 15]. Data was gathered in real time during 29 days with the aid of a MikroTik RB1100AHx router, and it created more than 867,000 rows of data with three logs per second. The logs were classified into three classes: Genuine, Fake, and No Heavy Activity. Pre-processing involved windowing sequences, normalisation, and SMOTE balancing, whereas the GRU model comprised update and reset gates, followed by a Dense layer and a Softmax 3-class output. The model was trained with categorical cross-entropy loss and optimized with the Adam optimizer, validated with a 5-fold cross-validation strategy. The results achieved a 86.8% mean accuracy and an F1 score of 0.90 in the classification of Genuine Bandwidth, indicating that the GRU can successfully detect temporal patterns in network logs. This system is locally deployable through the G-Radio interface, demonstrating its feasibility, scalability, and substantial contribution to automatic bandwidth classification without packet inspection.

Blockchain for Secure Electronic Health Records Management

2025-04-12T07:50:58+07:00

Electronic Health Records (EHRs) are essential to modern healthcare infrastructure, yet they face persistent challenges related to data security, interoperability, and unauthorized access. Blockchain technology, through its use of cryptographic protocols, smart contracts, and consensus mechanisms, offers a decentralized and tamper-resistant solution for managing EHRs. This paper explores the potential of blockchain in addressing critical limitations of conventional EHR systems, focusing on data immutability, fine-grained access control, and real-time data synchronization. By leveraging distributed ledger technology (DLT), the proposed approach reduces single points of failure and mitigates cybersecurity vulnerabilities. Furthermore, this study discusses practical implementations and case studies that demonstrate how blockchain can enhance trust, transparency, and efficiency in health data management. The analysis reveals that a well-designed blockchain-based EHR framework can minimize data breaches, protect patient data rights, and streamline operations across healthcare institutions. The paper concludes that blockchain presents a transformative path forward for secure and interoperable EHR systems. Future research should aim to refine architectural models, address scalability constraints, and ensure compliance with healthcare regulations and standards such as HIPAA and GDPR to facilitate real-world adoption and sustainable deployment.

DNA Sequence Classification Using Machine Learning Models Based on k-mer Features

2025-03-07T08:58:52+07:00

Cell-free DNA (cfDNA) has emerged as a promising biomarker in various clinical applications, particularly in cancer detection, prenatal diagnostics, and disease monitoring. Accurate classification of cfDNA sequences is crucial for improving diagnostic reliability and enabling timely clinical decisions. This study investigates the application of machine learning models—Decision Tree (DT), Support Vector Machine (SVM), and Deep Neural Network (DNN)—for classifying cfDNA sequences using k-mer-based feature extraction, with k set to 3. A total of 3,000 DNA sequences comprising both normal and tumor-derived samples were transformed into numerical feature vectors based on the frequency of 3-mer patterns. The models were trained and evaluated using standard metrics including accuracy, precision, recall, and F1-score. Experimental results demonstrate that the DNN model achieved the highest classification performance, effectively distinguishing between normal and tumor cfDNA. In contrast, the DT and SVM models exhibited relatively lower performance, particularly in identifying normal sequences. The study also addresses challenges such as class imbalance and limitations of simple k-mer representations. These findings highlight the potential of deep learning approaches in improving cfDNA sequence analysis and open avenues for future research using more complex models, larger datasets, and feature engineering techniques to enhance classification accuracy and clinical applicability.