Improving Loan Status Prediction Accuracy with Generative Adversarial Networks: Addressing Data Scarcity and Bias

Enas A.; Ahmed M.; Mazin Abed; Bourair

doi:https://doi.org/10.54216/JISIoT.130116

Journal of Cybersecurity and Information Management | 2024

Concealed Chosen Plaintext Attack on Multiple S-boxes Based Image Encryption

Chosen plaintext attacks (CPA) pose a significant security risk to encryption algorithms. However, it can be difficult to perform such an attack without direct access to the encryption process. This paper introduces a new cryptoanalysis method that uses hidden CPA to analyze image encryption schemes based on substitution boxes (S-boxes) Unlike traditional CPA methods, the proposed algorithm does not require that they can directly into the encryption process. Instead, a hidden attack vector is embedded in the natural host image to reduce the risk of attack detection. By asking the owner of the encryption algorithm to encrypt this encryption image and provide a cipher image, the input vector can be compared with its encrypted counterpart This can have an effective S-box and break encryption the algorithm, which does not interact directly with the encryption process. Experimental results demonstrate that the proposed method can completely recover cipher images in cascading S-box encryption schemes, regardless of the number of S-boxes used. Additionally, it conceals the CPA vector within the host image imperceptibly, achieving a high PSNR of 49.47 dB, indicating minimal visual distortion. Furthermore, our CPA significantly outperforms existing techniques in speed, recovering a  grayscale image in just 1.2 seconds. This method provides a simple yet effective cryptanalysis tool to evaluate the security of such image encryption schemes against CPAs.

groups

Ahmed Rabea mail -

Mohamed G. Abdelfattah mail -

Abeer T. Khalil mail -

Ali E. Takieldeen mail

link https://doi.org/10.54216/JCIM.140108

Volume & Issue

Vol. Volume 14 / Iss. Issue 1

Details open_in_new

Journal of Cybersecurity and Information Management | 2024

Fortifying Connected Vehicles Based Cybersecurity Measures for Secure Over-the-Air Software Updates

The emergence of connected vehicles has transformed the automotive sector by enhancing the vehicle’s functionality, efficiency, and safety. The performance and security of these vehicles significantly rely on the deployment of the over-the-air software update. However, the execution of OTA comes with many challenges, especially with regard to security vulnerabilities and risks. The current paper delves into the complexities of the secure OTA software update for connected vehicles addressing the most critical issues; authentication; encryption and integrity verification, and risk management. Through advanced cryptographic methodologies, stringent authentication processes, and secure communication channels, automotive manufacturers and other service providers can guarantee the integrity and confidentiality of the updates, and consumers’ data from malicious attack. Moreover, the paper explores the regulatory and other standards-related matters that control the use of OTA in the automotive sector. An understanding of the secure OTA update mechanisms aids the stakeholders in establishing a resilient connection in connected vehicles boosting consumer trust and the future of the automobiles industry.

groups

Shashikant Patil mail -

Senthil Kumar A. mail -

Saket Mishra mail -

N. Gobi mail -

Intekhab Alam mail -

Romil Jain mail

link https://doi.org/10.54216/JCIM.140109

Volume & Issue

Vol. Volume 14 / Iss. Issue 1

Details open_in_new

Journal of Cybersecurity and Information Management | 2024

Design of Novel Cryptographic Model Using Zero-Knowledge Proof Structure for Cyber Security Applications

Privacy and security in the current modern, digital communication and data transfer-oriented world has become imperative. Most commonly used encryption methods often involve exposing sensitive information, which might be an open gate for potential vulnerabilities. This paper aims to explore the topic of applying ZKPs in cybersecurity in a comprehensive manner. For this purpose, Proposed work will provide an exhaustive description of the basic concepts of Zero-Knowledge Proofs , which refer to both the interactive and non-interactive forms of the product. Additionally, the study will focus on presenting various cryptographic protocols and algorithms utilizing Zero-Knowledge Proofs , such as zk-SNARKs and zk-STARKs . In addition to theoretical studies, Proposed work analyze the practical implementation details of Zero-Knowledge Proofs implementations , cryptographic libraries, programming languages, and frameworks commonly used to create ZKP-based applications . Zero-knowledge proofs enable groundbreaking approaches to address cybersecurity problems with an emphasis on user privacy and data confidentiality. On average, cryptographic operations experienced delays of approximately 10 milliseconds which was not intrusive for real-time systems. The system’s throughout remained at a steady average of 100 Mbps all times, so it performed well at processing data despite cryptographic overhead. The packet delivery ratio was constantly high at 98%, implying that most data packets were delivered consistently even over encrypted communication paths.

groups

S. Anthoniraj mail -

Rahul Mishra mail -

Shweta Loonkar mail -

Trapty Agarwal mail -

Gunveen Ahluwalia mail -

Amandeep Gill mail

link https://doi.org/10.54216/JCIM.140110

Volume & Issue

Vol. Volume 14 / Iss. Issue 1

Details open_in_new

Journal of Cybersecurity and Information Management | 2024

Enhancing Energy Efficiency in Heterogeneous Cyber Security Networks Using Deep Q-Networks Data Routing

Since heterogeneous wireless sensor networks consist of sensor nodes of varying capacity and energy-constrained, effective routing techniques are essential to ensure the proper functioning of the systems. Most traditional routing techniques fail to dynamically adjust to varying network conditions, leading to ineffective use of energy and poor performance. Therefore, deep Q-Networks implementation of reinforcement learning provides a beneficial approach to the problem due to adaptive routing decisions depending on the environmental signals and systems’ performance. Therefore, the suggested approach integrates Deep Q-Network into the data routing framework for different Wireless Sensor Networks to improve energy-efficiency and ensure data delivery. The DQN agent is designed to pick routing functions that maximize total rewards which depend on energy consumption, packet delivery, and network stability. Hence, the decentralized learning allows each sensor node to develop its routing policy based on the local environment under the interactions with their neighbors. Therefore, the approach promotes the ability to adapt and learn, crucial for changing network conditions. Thus, extensive simulation was conducted to assess the applicability of the DQN-based routing for different WSNs, proving the significant reducing of energy consumption compared to traditional routing approaches with an average of 25% regardless of the network formation and traffic conditions . This approach also demonstrates lower packet loss of 15%, revealing enhanced data transfer reliability . In particular, the existing on demand routing protocols, only forward the request that arrives first from each route discovery process. The attacker exploits this property of the operation of route discovery. The network lifetime was extended by 30% showing growing energy efficiency for a long-term run. In general, the integration of Deep Q-Networks into data routing provides an excellent opportunity to improve energy-efficiency in different types of wireless sensor networks. Hence, the proposed approach effectively optimizes the routing solutions in real-time, using adaptive lenience, also showing enhancing data delivery, and improving the systems’ lifetime. Hence, the presented results prove the capability of reinforcement learning methods to address the growing challenges of WSNs and leave space for further research in autonomous WSN improvement.

groups

Gowrishankar J. mail -

Bhargavi Gaurav Deshpande mail -

Dhiraj Singh mail -

Awakash Mishra mail -

Zeeshan Ahmad Lone mail -

Bharat Bhushan mail

link https://doi.org/10.54216/JCIM.140111

Volume & Issue

Vol. Volume 14 / Iss. Issue 1

Details open_in_new

Journal of Cybersecurity and Information Management | 2024

Cyber Security Protection in Roadside Unit Based on Cross-Layer Adaptive Graph Neural Networks (Gnns) in Vanet

The proposed systems can improve cyber security in VANET applications by enabling efficient detection of complex attacks on the RSU component. The subsequent sections discuss the systems that are applied and support the suggestions for improving the VANET trustworthiness. VANETs and show that the utilization of Cross-Layer Adaptive GNNs can improve cyber security and LEARNING in VANET-based RSUs. As a result, the suggested system can provide robust ways for detecting cyber-attacks by: modeling the network architecture using graphs while combining information regarding several protocol layers to detect complicated interactions between the network entities and find the abnormal activities. the nature of the GNN enables it to update in real-time by adapting to the evolving attack patterns and the shifting network conditions, making them sturdy and flexible defense ways for cyber security. The proposed network e systems can efficiently detect multiple cyber threats and focus on reducing the number of false positives while maintaining low computation costs. Therefore, chances are that incorporating the Cross-layer adaptive GNNs into the RSUs can improve cyber security in VANETs, enhancing the robustness and reliability of prospective smart transportation systems.

groups

Raj Kumar mail -

Sakshi Pandey mail -

Asha KS mail -

Rakesh Kumar Yadav mail -

Abhinav Mishra mail -

Sunil Sharma mail

link https://doi.org/10.54216/JCIM.140112

Volume & Issue

Vol. Volume 14 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2024

Optimizing Brain Tumor Classification Accuracy Through Transfer Learning and Internet of Things Integration

Brain tumor classification using medical images is crucial for identification and therapy. However, brain tumors are complex and vary, making grouping them difficult. This work demonstrates a novel transfer learning method for brain tumor classification. We employ trained Convolutional Neural Networks (CNNs) models and data enrichment approaches to extract meaningful information from medical images. We want to fine-tune the models built on our dataset to uncover hierarchical patterns that distinguish tumor types. Through data enrichment, the training sample becomes more diverse and richer, making the model more generic and robust. Our team's extensive testing and research have shown that the suggested procedure can identify brain tumors. Our machine-learning approach performs better than others in terms of accuracy, sensitivity, specificity, and precision. Our technique improves brain tumor categorization and assures accurate clinical diagnosis. Automated testing systems are one way for physicians to assist patients in selecting the best course of treatment. Researchers may improve classification performance by incorporating modern imaging technology or topic-specific data. The Internet of Things, or IoT, is helping to drive the development of complex real-time data collection, processing, and sharing systems. These technological advancements have transformed medical imaging. This graphic depicts a cutting-edge transfer learning system that may be able to identify brain cancer from medical photos. This technology has the potential to enhance data collection and processing via the Internet of Things. Data augmentation and pre-trained convolutional neural networks may help to extract interpretable medical images. The Internet of Things improved the model's flexibility, resilience, and utility. We achieved this by expanding the training data set. Rapid categorization advancements have made clinical diagnosis more efficient. Classification, deep learning, medical imaging, machine learning, transfer learning, tumor detection, and image analysis all relate to this topic.

groups

Bhanu Bhushan Parashar mail -

Munesh Chandra mail -

Sachin Malhotra mail

link https://doi.org/10.54216/JISIoT.130112

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2024

An Advance Study of an Efficient CNN-Grounded Deep Learning Classification Technique for the Diagnosis of IoT based Cardiac Arrhythmias

Deep Learning, or DL for short, is an emerging subfield within the larger discipline of machine learning in today's world. The study being conducted in this area is progressing at an immediate stride, and the discoveries are contributing to the progression of technology. Deep learning (DL) methods were developed with the intention of developing a general-purpose learning method that would enable the gradual learning of characteristics at multiple levels without relying on human-engineered features. This was the goal of deep learning. Because of this, the system is able to acquire intricate purposes and directly map input to output by making use of the data that it has acquired which is based on Internet of things (IoTs). This study places an emphasis on the application of CNN (Convolutional Neural Networks), which are a subcategory of DNN (Deep Neural Networks), and it develops an efficient layered CNN for the classification of ECG arrhythmias. Even while FC-ANNs (Fully Connected Artificial Neural Networks), which are sometimes referred to as Multilayer-Perceptron networks, are effective in categorising ECG arrhythmias, the optimization process for many classification networks takes a significant amount of time in terms of computation. In addition, the features extracted by engineers are what define the accuracy of the categorization of ECG arrhythmias. An improved CNN based filtering, feature abstraction, and classification prototypical is established in order to conduct an accurate analysis of an electrocardiogram (ECG). When measured against ANN, the performance was found to have an accuracy rating of 99.6%. Consequently, the CNN model that was suggested is useful to doctors in arriving at the definitive diagnosis of AFL (atrial flutter), AFIB (atrial fibrillation), VFL (ventricular flutter), and VT (ventricular tachycardia). It includes denoising, feature extraction, and categorization as part of its functionality.

groups

Deepa devasenapathy mail -

Rohit Pachlor mail -

Ramesh M. mail -

G. Shanmugaraj mail -

Aby K. Thomas mail -

K. Sridhar mail

link https://doi.org/10.54216/JISIoT.130113

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2024

An Ensemble Machine Learning Method for Analyzing Various Medical Datasets

In recent years, machine learning (ML) has shown a significant impact in tackling various complicated problems in different application domains, including healthcare, economics, ecological, stock market, surveillance, and commercial applications. Machine Learning techniques are good enough to deal with a wide range of data, uncover fascinating links, offer insights, and spot trends. ML can improve disease diagnosis accuracy, predictability, performance, and reliability. This paper reviews various machine learning techniques applied to different medical datasets and proposes an ensemble method for helping in the early diagnosis of different diseases. The study compares existing machine learning techniques with the proposed ensemble method. The ensemble method uses the AdaBoost algorithm to combine the traits of choice trees, random forests, and support vector machines. Three feature selection techniques, Fisher’s score, information gain, and genetic algorithm, are used to select appropriate dataset features. The ensemble method also uses the K-fold cross-validation technique (where k=15) for validating results. SMOTE was employed to balance some of the datasets because they were quite unbalanced. All the methods used in this study are evaluated based on accuracy, AU Curve, Recall, Precision, and F1-score. The paper uses different medical datasets at the University of California Irvine and the Kaggle directory to compare machine-learning models with the proposed ensemble method. The encouraging results show that the ensemble method outperforms the existing machine-learning techniques. The paper thoroughly analyzes how machine learning is used in the medical industry, covering established technologies and their impact on medical diagnosis. An early diagnosis is needed to prevent people from deadly diseases. Hence, this study proposes an ensemble method that may be used to diagnose different diseases early.

groups

Chhaya Gupta mail -

Nasib Singh Gill mail -

Priti Maheshwary mail -

Shraddha V. Pandit mail -

Preeti Gulia mail -

Piyush Kumar Pareek mail

link https://doi.org/10.54216/JISIoT.130114

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2024

Systematic Analysis based on Conflux of Machine Learning and Internet of Things using Bibliometric analysis

IoT devices produce a gigantic amount of data and it has grown exponentially in previous years. To get insights from this multi-property data, machine learning has proved its worth across the industry. The present paper provides an overview of the variety of data collected through IoT devices. The conflux of machine learning with IoT is also explained using the bibliometric analysis technique. This paper presents a systematic literature review using bibliometric analysis of the data collected from Scopus and WoS. Academic literature for the last six years is used to explore research insights, patterns, and trends in the field of IoT using machine learning. This study analyses and assesses research for the last six years using machine learning in seven IoT domains like Healthcare, Smart City, Energy systems, Industrial IoT, Security, Climate, and Agriculture. The author’s and country-wise citation analysis is also presented in this study. VOSviewer version 1.6.18 is used to provide a graphical representation of author citation analysis. This study may be quite helpful for researchers and practitioners to develop a blueprint of machine learning techniques in various IoT domains.

groups

Ayushi Chahal mail -

Santosh Reddy Addula mail -

Anurag Jain mail -

Preeti Gulia mail -

Nasib Singh Gill mail -

Bala Dhandayuthapani V. mail

link https://doi.org/10.54216/JISIoT.130115

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new

Journal of Intelligent Systems and Internet of Things | 2024

Improving Loan Status Prediction Accuracy with Generative Adversarial Networks: Addressing Data Scarcity and Bias

A precise and reliable loan status prediction is of the essence for financial institutions, However, the lack of real-world data and biases within that data can greatly impact the accuracy of machine learning models. Another challenge faced by loan status prediction models is class imbalance, where one category (such as approved loans) is much more common than another (such as defaulted loans), leading to skewed predictions towards the majority class. This study inspects Generative Adversarial Networks (GANs) to augment the data and improve the machine learning models’ performance. Several machine learning (ML) models including but not limited to Support Vector Machines (SVM) and ensemble bagged trees were employed on a Kaggle loan dataset (380 samples). Baseline training and testing accuracies were 86.9% and 86.3% (SVM) and 84.5% and 82.1% (ensemble). ActGAN (Activating Generative Networks) was then utilized to generate synthetic data points for both accepted and rejected loans. Retraining the models with new augmented data showed remarkable improvements: SVM accuracies for training and testing rose to 94.4% and 93.4%, while ensemble models achieved 97.4% and 95.8%, respectively. Other ML models were also explored such as KNN, Decision tree and logistic Regression and showed promising results in terms of accuracy as compared to the state of art. These findings put forward that GAN-based data augmentation can enhance the performance of loan status prediction. Future research could explore GAN’s impact of different architectures and assess the general applicability of this approach.

groups

Enas A. Raheem mail -

Ahmed M. Dinar mail -

Mazin Abed Mohammed mail -

Bourair Al-Attar mail

link https://doi.org/10.54216/JISIoT.130116

Volume & Issue

Vol. Volume 13 / Iss. Issue 1

Details open_in_new