Journal of Intelligent Systems and Internet of Things

One of the most effective devices to model uncertainty in decision-making difficulties is the neutrosophic set (NS) and its extensions, namely interval NS (INS), interval complex NS (ICNS), and complex NS (CNS). An effective device to demonstrate ambiguities and uncertainty in decision-making is the NS, which is the more conventional standard set, intuitionistic fuzzy set (IFS), and fuzzy set (FS) by including 3 scores of falsehood, indeterminacy, and truth of established statements. Financial risk management is a massive field with different and developing modules, as demonstrated by either its historic growth or present classic example. It is a procedure to address the uncertainty originating from financial markets. It consists of calculating the financial threats dealing with organization and emerging management tactics by internal policies and priorities. A risk-management method is an experience control and accounting system. In this manuscript, we develop an Intelligent Risk Management Approach for Financial Crisis Using Pythagorean Neutrosophic Fuzzy Graphs and Metaheuristic Optimization Algorithms (IRMFC-PNFGMOA). The main intention of IRMFC-PNFGMOA technique is to analyse and develop effective methodologies for measuring and managing financial risk in dynamic market conditions. Initially, the data pre-processing stage applies Z-score normalization to clean, transform, and structure raw data to improve the quality. Besides, the Aquila optimization algorithm (AOA) has been deployed for the selection of feature processes to identify and retain the most relevant features from input data. For the classification process, the proposed IRMFC-PNFGMOA model designs pythagorean neutrosophic fuzzy graphs (PNFG) technique. To further optimize model performance, the growth optimizer (GO) algorithm is utilized for hyperparameter tuning to ensure that the best hyperparameters are selected for enhanced accuracy. To exhibit the enhanced performance of the presented IRMFC-PNFGMOA model, a comprehensive experimental analysis is made. The comparative results reported the improvised characteristics of the IRMFC-PNFGMOA model.

The research evaluates the effects, which cloud computing and digital educational methods have on scholarly performance. The research used descriptive statistics combined with t-tests alongside ANOVA and regression analysis for interpreting the data findings. The collected data shows students use cloud computing moderately and employ digital education extensively although their educational outcomes stay average. Cloud computing usage exhibited similar levels of acceptance between male and female students however, students from arts streams programs demonstrated increased interest. Cloud computing usage along with digital education experienced superior adoption rates among students residing in rural areas than students settled in urban areas. Research data showed a major statistical linkage between digital education and the levels of academic performance. The educational institution types together with parental work status shaped student interaction with digital educational resources. The study's findings highlight the significant roles played by cloud computing and online learning in raising students' academic performance. The research implies that mixing technology with current education practices will boost educational results while demonstrating why digital competence stands vital in present-day education systems. Academic achievement rates improved in direct proportion to the amount of digital education use by students alongside the fact that private institution students demonstrated higher application of cloud computing platforms and female students demonstrated superior academic outcomes when compared to male students. Numerous students adopt both cloud computing systems and digital education methods because such technology usage is prevalent at accuracy 91.4% of the total students. Out of all the analyses done in the research, the overall F1-score is 92.5, and the fault tolerance of 93.8%.

The diagnosis of breast cancer depends on histopathology for precise and trusted evaluation between malignant tumor cells and benign cells. The analysis demands significant time and creates additional room for human errors. A deep learning approach for computer-aided diagnosis (CAD) establishes techniques to enhance the classification performance in this study. The proposed methods utilize One-hot encoding with VGG-16 for feature extraction to achieve 98% accuracy with BreakHis data while DBN for feature learning reaches 98% accuracy on BreakHis and 96% on Kaggle. SSGAN addresses unannotated images effectively with up to 89% accuracy. Through its application, deep learning technology proves to enhance breast cancer identification while decreasing the workload on medical pathologists. One-hot encoding remains efficient for computations yet the DBN extraction method produces superior features. The SSGAN model increases labeling accuracy when it uses available labeled data and unlabeled data to lower annotation expenses. Deep learning technologies validate their ability to transform breast cancer histopathological diagnosis through precision-enhanced efficient examination methods especially with semi-supervised GAN systems.

Spatio-temporal human activity recognition (HAR) is an emerging field that uses spatial and temporal data to identify and classify human activities accurately. It has been effectively applied in areas like healthcare for monitoring daily activities, detecting anomalies, and aiding rehabilitation with real time feedback. However, there is a gap in research specifically focusing on integrating spatio temporal data with advanced machine and deep learning techniques for HAR based on sensor data. Existing reviews do not comprehensively cover spatio-temporal HAR based on sensor data, resulting in a lack of summaries on recent models, datasets, sensor technologies, applications, and machine/deep learning techniques used in this field. This systematic review provides a comprehendsive overview of spatio-temporal HAR based on sensor data, tracing its development from the origin of sensor-based spatio-temporal HAR field to the present. It highlights the main challenges in spatio- temporal HAR. The review also examines model trends over the years, including the distribution of models used in HAR and the identification of those frequently combined to form hybrid models. Additionally, it analyzes accuracy trends of the commonly used datasets and identifies the datasets that are widely used in spatio-temporal HAR research. Furthermore, various application domains and sensor technologies used in spatio-temporal HAR are identified.

The primary objective of designing routing protocols for Wireless Sensor Networks (WSNs) is to extend the network lifetime by optimizing the use of the limited battery energy of the sensor nodes. To improve conservation of energy and longevity of the network in WSNs, this study proposes a Cluster-based Chain-Tree Routing Protocol (CCTRP). Integrating tree based chain and cluster routing methods in WSNs is the primary objective of this study. This new CCTRP adopts a sector-based vertical network-partitioning scheme that divides network into sectors and it again vertically partitions the nodes too form various size of clusters. Then, Minimum Spanning Tree (MST) is created based on the kruskal’s Algorithm through a Chain Leader (CL) node serving as the receiver and chain is formed from CLs of last level cluster to Base Station (BS) in each sector. Using the BS, remaining energy and distance to the next CL node, CCTRP determines the Cluster Leader (CL) or Chain CL node in each cluster. For data transport, it also selects the shortest paths. When the energy that remains in the node is ready to be exhausted, the transition is executed according to this protocol. This results in a significant improvement of the average network lifespan. Finally, the CCTRP protocol outperforms the current protocols in terms of network performance, according to the simulation results.

Underwater optical communication (UOC) and off-surface areas wireless communications are a rapidly growing field, especially with the emergence of new technologies such as autonomous underwater vehicles and above/water drones. The challenge lies in the absence of a water surface platform to transfer the signal from underwater to off surface. This research investigates the design and implementation of a hybrid communication system that successfully transmits signals from underwater environments to above-water. The study utilizes OFDM as method to generate data on the integration of underwater optical wireless communication (UWOC) at 532nm and LOS optical channel. After adjusting the line of sight through the angle of refraction and overcoming the challenges of water and above water conditions as well as ambient lighting, ambitious results were obtained 100 meters above clear water and 40 meters in haze wither at a depth of 10 meters for transmission. The research has mitigated challenges and enhancing the effectiveness of underwater-to-air communication systems.

Early detection of plant diseases is critical to minimizing their adverse effects on agricultural productivity. In particular, machine vision and deep learning approaches (e.g., convolutional neural networks, CNNs) have been increasingly applied for automatic plant disease identification. Although existing deep learning models achieve satisfying classification accuracy, they often consist of millions of parameters that significantly lead to the lengthy training time, prohibitive calculation costs and deployment obstacles at the resource-constrained edge devices. In order to overcome those constraints, we introduce a new deep learning architecture, which uses adaptations of Inception layers and residual connections that can help both with feature extraction and efficiency. In addition, depthwise separable convolutions are used to drastically reduce the amount of trainable parameters with small loss of representational power. We perform training and evaluation of the proposed model on three located benchmark plant disease datasets, PlantVillage dataset, the Rice Disease dataset. Experimental results show that our model achieves state-of-the-art classification accuracy of 99.39% on the PlantVillage dataset, 98.66% on the Rice Disease dataset. In contrast to the state-of-the-art deep learning models, our method obtains higher accuracy with fewer parameters so that it could be better suited for real-time applications on mobile and embedded systems. We explore an application of deep learning with the use of optimized architectures and present the viability of this technique in precision agriculture for faster and more accurate diagnosis of diseases in plants with lower computational load.

  Data security, privacy, sensitivity, and integrity are major concerns when using cloud-based storage solutions. In this paper, we propose a hybrid encryption model that has been integrated with blockchain technology to secure data storage in the cloud. The proposed model facilitates data encryption using a symmetric cryptography algorithm for efficient large data encryption while ensuring the encryption key can only be exchanged using asymmetric cryptography. This model utilizes the power of blockchain to manage metadata securely and associated encryption keys to ensure that records are tamper-proof, removing the need for third parties to be trusted. The security, key management, and data integrity of the proposed model are better than traditional cloud storage and existing blockchain-based approaches. The performance evaluation suggests that the model achieves a balance between security and cost efficiency, while moderate transaction speed will be witnessed owing to blockchain operations. Our proposed work aims to provide a scalable, fast, reliable, and decentralized architecture-based solution to address the challenges of cloud data security.  

This study investigates combining fuzzy logic with deep learning methodologies in classifying X-ray images for osteoporosis detection. Osteoporosis, defined by compromised bone integrity and heightened fracture susceptibility, requires prompt and precise diagnosis for effective treatment. We devised a hybrid approach that amalgamates transfer learning from Convolutional Neural Network (CNN) architectures, including MobileNetV2, AlexNet, ResNet50V2, and Xception, utilizing fuzzy logic during the preprocessing phase to address uncertainty and imprecision in X-ray images, thereby enhancing the quality of the input data for the subsequent pre-trained models. The research entailed the examination of a significant dataset of X-ray images and the implementation of the proposed methodology to categorize images as osteoporotic or non-osteoporotic, attaining a remarkable accuracy of 99.68% and a receiver operating characteristic (ROC) of 100% through the integration of fuzzy logic preprocessing with ResNet50V2. This innovative method may substantially decrease diagnostic inaccuracies and enhance patient outcomes, facilitating additional research and development in applying deep learning techniques in healthcare.

This study proposes a deep learning-based framework to detect and classify Alzheimer's disease (AD) in the early stages using medical imaging, and specifically Magnetic Resonance Imaging (MRI). Specifically, we propose a Convolution Neural Network (CNN) based model and transfer learn (MobileNet) through pre-trained models based on task domain to improve model performance on binary AD classification. Thanks to minimizing computational complexity and memory costs, the model with 99.86% accuracy rate can mitigate overfitting and is an ideal approach for real time and eco-friendly monitoring of AD evolution. The findings suggest that the model could help clinicians in diagnosing AD even based on MRI images, which has great potential as a scalable and efficient solution for the early-stage diagnosis and classification of the disease. Our work will include the addition of further pre-trained models, increased dataset size via data augmentation, and the application of MRI segmentation to better isolate some of the key features of Alzheimer.

Inventory management is crucial for optimizing consumer demand and supply chains in e-commerce companies.  This is the stage at which precise inventory forecasting becomes necessary for forecasting future demand patterns and stock levels.  Traditional forecasting methods often struggle with e-commerce data due to seasonality, sudden changes in customer behavior, and nonlinearity.  Machine learning (ML) and deep learning (DL) techniques have become powerful weapons for inventory prediction because they can analyze huge amounts of data with high dimensionality. E-commerce firms can improve their resource allocation, inventory management, and customer experience in highly competitive market environments.  This paper proposes different types of inventory forecasting models and mainly evaluates the applicability of sophisticated machine learning algorithms.  While we commonly use old methods like Random Forest, ARIMA, and MLPs, they often lack the necessary robustness to nonlinearity within inventory data.  To address these problems, we introduce a novel method that combines convolutional neural networks (CNN) and XGBoost called CNN-XGBoost, which provides better feature extraction than the conventional prediction model and regression performance.  We then compared CNN-XGBoost's performance to traditional forecasting methods (another common approach to contextualizing predictive model performance) using a 52-week simulated dataset in which we mimic patient data growing over time.  We used key performance metrics such as R2, mean squared error (MSE), and mean absolute percentage error (MAPE) to assess each model's accuracy.  The CNN-XGBoost model performed much better than others, with an R2 of 0.78, which means our proposed model can explain more variation compared to other competitors, as depicted in the results section.  It also had the best MSE of 0.15, indicating better predictive performance.  The CNN-XGBoost model demonstrated promising prospects as a robust inventory forecasting tool for commerce despite its slightly higher MAPE value (0.69), suggesting some vulnerability to outlier data points.  This study demonstrates the potential of using a convolutional neural network in combination with gradient boosting techniques to tackle the complexity of stock management issues and the fact that it outperforms based line methods by a large margin.

The Internet of Things (IoT) has extensively converted the industry of tourism, reforming travel design, supply, and experiences. The technology of Blockchain (BC) signifies a paradigm shift with the latent to transform many industries, more like spreadsheets altered office efficiency. BC technology provides frequent potential advantages to the tourism industry, with enhanced transparency, security, and efficacy in regions such as payments, bookings, and identity verification, which potentially mains to a more perfect and reliable travel experience. In the tourism region, BC with IoT is mainly attractive owing to the latent benefits it provides in terms of improving the experience of tourism, enhancing operational efficacy, and guaranteeing data security and transactions. Recently, numerous scholars globally have employed deep learning (DL) technology in the industry of tourism to combine physical and social influences for improved travel recommendation services. This study presents a Blockchain for Tourism Service Customization and Management using Whale‐goshawk Optimization Algorithm (BCTSCM-WOA) technique. The main goal of the BCTSCM-WOA method relies on improving the effectual model for tourism service customization. Initially, blockchain technology is applied to provide secure, transparent, and decentralized solutions for handling traveler data, payments, and service personalization. Then, the data pre-processing employs min‐max scaling to transform input data into a suitable format. Besides, the crayfish optimization algorithm (COA) to select the most relevant features from the data has executed the feature selection procedure. For the classification process, the proposed BCTSCM-WOA method projects multi-dimensional attention-spiking neural network (MASNN) technique. At last, the parameter tuning process is performed through the whale‐goshawk optimization (WGO) algorithm for refining the classification performance of MASNN model. The experimental evaluation of the BCTSCM-WOA algorithm has been examined on a benchmark dataset. The extensive outcomes highlight the significant solution of the BCTSCM-WOA approach to the classification process when compared to existing techniques.

Diabetic Retinopathy (DR) is a general difficulty of diabetes mellitus, resulting in retina damage that affects vision. If left undetected, it has the potential to cause blindness. Regrettably, DR is irreversible, and only treatment can maintain vision. The early analysis and treatment of DR can considerably decrease the potential for visual impairment. Unlike computer-aided diagnosis (CAD) systems, the manual diagnostics method of DR retinal images by ophthalmologists is effort-, cost-, and time-consuming and liable to misdiagnoses. In present scenario, deep learning (DL) has become the classical approach that has remarkable performance in different fields, mainly in medical image classification and analysis. Convolutional neural networks (CNN) are more commonly deployed as a DL system in medical image analysis and they are very efficient. In this manuscript, we offer the design of Tent Chaotic Dung Beetle Optimization with Deep Ensemble Learning for Diabetic Retinopathy (TCDBO-DELDR) Recognition approach on Fundus Imaging. The foremost intention of the TCDBO-DELDR technique is to automate the DR detection process on fundus images via the ensemble DL model. To eradicate the noise, the TCDBO-DELDR technique initially exploits the median filtering (MF) methodology. In the TCDBO-DELDR model, the Inception v3 (IV3) model is employed for the purposes of feature extractor. For the hyperparameter tuning procedure, the TCDBO technique is used for IV3 model. Finally, the detection of DR is carried out utilizing an ensemble of three classifiers namely Deep Feedforward Neural Network (DeepFFNN), Convolutional FFNN (ConvFFNN), and Convolutional bi-directional long short-term memory (ConvBLSTM). For ensuring the enhanced efficiency of the TCDBO-DELDR system in the DR detection procedure, a widespread experimental study is prepared on the benchmark DR database. The results illustrate the superior efficiency of the TCDBO-DELDR technique with other recent DL approaches.

The express expansion of the Internet of Things (IoT) has led to an exponential increase for data being generated and transmitted from various connected devices. This poses significant challenges in terms of data privacy and security, as unauthorized access to such sensitive information can have severe consequences like identity theft or financial fraud. This research proposes a model for sensitive data detection and protection in IoT, based on deep learning and optimization-enabled secure encryption. By combining deep learning-based sensitive data detection and optimization-enabled secure encryption, this model offers a comprehensive solution to preserve data privacy in IoT. The proposed model uses a novel and secure encryption algorithm, ensuring the privacy of the data. An algorithm, Improved Skill Optimization Algorithm (ISOA), which enhances the performance of existing optimization algorithms by incorporating the concept of Double Exponential Smoothing (DES), is proposed for the secure key generation for the data encryption. Data Encryption Standard (DES) is a block cipher algorithm that encrypts and decrypts data using a 56-bit key and 64-bit blocks. The proposed model provides a robust solution for data privacy preservation in IoT networks, which is crucial for protecting sensitive information from unauthorized access and data breaches. The proposed algorithm's performance analysis is evaluated using metrics, like computation time, memory, and fitness function. Results indicate that proposed ISOA based encryption model succeeded a greater performance, with a memory of 0.5170 MB, computational time of 1126.47 sec and fitness value of 1.3630.

Diabetes mellitus remains a global health concern, necessitating both accurate and effective diagnostic methodologies. This condition presents a significant challenge due to the high dimensionality of clinical datasets and the inherent complexity of diabetes classification. To address this problem, this study integrates feature selection and machine learning architectures to enhance diabetes prediction accuracy. A novel framework based on the Binary Greylag Goose Optimization (bGGO) algorithm is proposed to optimize feature selection, thereby improving classification performance. A comprehensive evaluation uses multiple classifiers, including Decision Trees, k-nearest Neighbors, Support Vector Machines, Random Forests, and Multilayer Perceptron (MLP). The experimental results demonstrate that bGGO significantly enhances feature selection quality, improving classification metrics, particularly for MLP, which achieves the highest classification accuracy of 95.98%. These findings underscore the efficacy of combining metaheuristic optimization with machine learning for diabetes diagnosis, offering a scalable and interpretable approach for real-world healthcare applications. The proposed methodology contributes to more precise risk estimation and the development of individualized intervention strategies, facilitating early diagnosis and effective disease management.