The escalating proliferation of devices, including both Internet of Things (IoT) and non-IoT devices, has triggered a suite of emergent security challenges in cyberspace, such as accurate device identification and authentication. The wide array of device types, protocols, and usability exacerbates these challenges. While conventional addressing schemes such as the logical Internet Protocol addressing and physical Media Access Control addressing schemes are integral for communication, they are susceptible to spoofing attacks. Device fingerprinting can be used to address the issue of identifying devices and traffic types using only implicit identifiers such as network traffic characteristics. In this paper, supervised machine learning based a device fingerprinting model has been proposed for the classification of both IoT and non-IoT devices on three levels based on their communication traffic characteristics. A meticulous feature selection process, employing two attribute evaluators, identified a subset of twenty features crucial for generating unique fingerprints from a large set of features pool. Three publicly available datasets and two supervised classifiers were utilized for evaluation purposes. Experimental results illustrated that the proposed model attained a classification accuracy exceeding 99% in discerning between known and unknown traffic traces (Level-1) on both the UNSW IoT and D-Link IoT datasets using the Random Forest (RF) classifier, and 99.74% accuracy in classifying network traffic types (Level-2) on the UNSW dataset. Individual device identification (Level-3) proves equally robust, with the RF and J48 classifiers achieving 99.03% and 98.14% accuracies on the UNSW non-IoT and IoT datasets, respectively. These findings underscore the potential of the device fingerprinting model in enhancing network security. The model’s robust classification capabilities across various datasets and identification levels make it a valuable asset in tackling modern security challenges in networked environments.

The movement of capital, integration, distribution, and social supply and demand adjustment are all greatly aided by commercial banks; yet, integrating credit risk assessment is a difficult task for banking Human Resource Management (HRM). To overcome these issues, a novel credit risk assessment in HRM frameworks is done using the Enhanced Humboldt Squid based Probabilistic Spiking Neural Networks with Shunted self-attention (EHSPNN-SSA) method is proposed. Initially, the input commercial bank datasets are taken from General Data Protection Regulation (GDPR) and Advanced Analytics of Credit Registry (AACR) Datasets. Then these data are pre-processes using Grid-Restricted Data Filtering Approach (GRDFA). After that, the data is extracted using Hybrid Absolute deviation factors (ADFs) class document frequency (CDF) (hyb ADF-CDF) based feature extraction method. Then these data are classified using Enhanced Probabilistic Spiking Neural Networks with Shunted self-attention (EPSNN-SSA) and optimized using the Humboldt Squid Optimization Algorithm (HSOA). The framework is validated using real-world banking data and compared to existing methods to demonstrate its efficacy in assessing credit risk and optimizing human resource management processes. The results show that the introduced approach performs better than previous approaches in a number of performance measures, including risky data accuracy (99.6%), non-risky data accuracy (99.7%), and risky data accuracy (99.4%) for dataset 1 and dataset 2, respectively. This indicates the method's exceptional effectiveness and room for advancement in the field.

This paper presents the implementation and evaluation of a Multi-key Multi-modalities Biometric Encryption System designed for business enterprises, leveraging cloud storage for secure and scalable data management. The system integrates multiple biometric modalities fingerprint, iris scan, and face recognition to enhance data security through advanced multi-key encryption techniques, utilizing algorithms such as Advanced Encryption Standard (AES) and Rivest-Shamir-Adleman (RSA). The encrypted biometric data is securely stored in the cloud, providing enterprises with efficient storage solutions. The system's performance was evaluated across several parameters including encryption/decryption time, biometric match accuracy, data transfer speeds, energy consumption, cost, and user satisfaction. The results demonstrate that multi-modal systems offer superior accuracy and security compared to single-modality systems, reducing error rates and enhancing reliability. However, multi-modal authentication incurs higher costs, energy consumption, and slightly longer processing times. Despite these trade-offs, the system achieved high user satisfaction, particularly in high-security environments where data protection is a priority. The findings indicate that the proposed system is a viable solution for businesses seeking a secure, scalable, and efficient method of protecting sensitive data.

This study introduces an improved BERT-based model for sentiment analysis in several languages, specifically focusing on analyzing e-commerce evaluations written in English and Bengali. Conventional sentiment analysis techniques frequently face difficulties in dealing with the subtle linguistic differences and cultural diversities present in datasets containing multiple languages. The model we propose integrates sophisticated methodologies and utilizes Local Interpretable Model-agnostic Explanations (LIME) to enhance the accuracy, interpretability, and dependability of sentiment assessments in various language situations. To tackle the challenges of sentiment categorization in a multilingual setting, we enhance the pre-trained BERT architecture by incorporating extra neural network layers. Compared to traditional machine learning and current deep learning methods, the model underwent a thorough evaluation, showcasing its superior capabilities with accuracy, precision, recall, and F1-score of 0.92. Including LIME improves the model’s transparency, allowing for a better understanding of the decision-making process and increasing user confidence. This research highlights the potential of utilizing advanced deep learning models to address the difficulties of sentiment analysis in global e-commerce environments, providing major implications for both academic research and practical applications in industry.

The Routing Protocol for Low-Power and Lossy Networks (RPL) is a widely adopted protocol for managing and optimizing routing in resource-constrained Internet of Things (IoT) environments.  RPL operates by constructing a Destination-Oriented Directed Acyclic Graph (DODAG) to establish efficient routes between nodes. This protocol is designed to address the unique challenges of IoT networks, such as limited energy resources, unreliable wireless links, and frequent topology changes. RPL's adaptability and scalability render it particularly suitable for large-scale IoT deployments in various applications, including smart cities, industrial automation, and environmental monitoring. However, the protocol's vulnerability to various security attacks poses significant threats to the reliability and confidentiality of IoT networks. To address this issue, a novel deep-stacked neuro-fuzzy system (DSNFyS) has been developed for attack detection in RPL-based IoT. The proposed approach begins with simulating RPL routing in IoT, followed by attack detection processing at the Base Station (BS) using log data. Data normalization is accomplished through the application of min-max normalization techniques. The most crucial features are then identified through feature selection, utilizing information gain and Support Vector Machine-Recursive Feature Elimination (SVM-RFE). Attack detection is subsequently performed using DSNFyS, which integrates a Deep Stacked Autoencoder (DSA) with an Adaptive Neuro-Fuzzy Inference System (ANFIS). Upon detection of an attack, mitigation is carried out employing a DSA trained using the Hiking Optimization Algorithm (HOA). The proposed DSNFyS demonstrated exceptional performance, achieving the better accuracy of 97.41%, True Positive Rate (TPR) of 97.60%, and True Negative Rate (TNR) of 97.12%.

This paper conducts a comparative analysis of three widely adopted Agile methodologies: Scrum, Kanban, and Extreme Programming (XP). By examining their application across diverse software development environments, the study highlights each methodology's inherent strengths and explores their practical implications for managing complex, large-scale projects. Central to this investigation are the scalability challenges that become particularly pronounced in settings with extensive stakeholder groups and complex coordination needs. The research draws upon a robust literature review and case studies to identify these challenges, setting the stage for a discussion of innovative solutions aimed at refining Agile practices. While specific solutions are reserved for detailed treatment in the proposed solutions section, the abstract is written to underscore the critical need for scalable strategies that can adapt to the dynamic landscapes of modern project management. This comparative inquiry not only enriches the academic discourse on Agile methodologies but also serves as a vital resource for practitioners seeking to optimize their project management strategies in complex scenarios.

This paper presents the development of an intelligent information system for analysing the happiness index and life satisfaction based on sociological survey data from various countries. The research addresses the need to improve the accuracy and efficiency of social research by integrating data mining and machine learning methods – specifically K-means clustering and multiple regression analysis – into the system design. The proposed module enables automated classification of countries and cities by life satisfaction levels, allowing stakeholders to make informed decisions on urban planning and social policy. The system also facilitates the identification of favourable living environments, providing valuable insights into the social, economic, and environmental factors affecting well-being. The experimental results on real-world datasets confirm the module’s effectiveness and predictive capabilities.