Search

In the rapidly evolving landscape of the Internet of Things (IoT), ensuring robust intrusion detection is paramount for device and data security. This paper proposes a novel method for intrusion detection in IoT networks that leverages a unique blend of subspace clustering and ensemble learning. Our framework integrates three innovative strategies: Clustering Results as Features (CRF), Two-Level Decision Making (TDM), and Iterative Feedback Loop (IFL). These strategies synergize to enhance detection performance and model robustness. We employ mutual information for feature selection and utilize four subspace clustering algorithms – CLIQUE, PROCLUS, SUBCLU, and LOF – to create additional feature sets. Three base learners – NB, LGBM, and XGB – are used in conjunction with a Logistic Regression (LR) meta-learner. To fine-tune our model, we apply Particle Swarm Optimization (PSO) for hyperparameter optimization. We evaluate our framework on the UNSW-NB15 dataset, which contains realistic and diverse IoT network traffic data. The results show that our framework outperforms the state-of-the-art methods in terms of accuracy (97.05%), precision (96.33%), recall (96.55%), F1-score (96.45%), and false positive rate (0.029). Our framework can effectively detect both known and unknown attacks in IoT networks and achieve high accuracy and low false positive rate. The paper contributes both practical implications for network security and theoretical advancements in intrusion detection research. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) consists of heterogeneous devices such as wearables and monitoring devices that collect data to provide autonomous decision-making and smart applications. IoT technologies, such as the Internet of Medical Things (IoMT), have become gradually popular for medical purposes, combining IoT and medical devices to achieve good health and well-being. However, IoMT devices are often tight and have resource constraints, which leads to limited local data processing in the device. Edge computing provides access to additional computation and storage resources for IoMT devices, bringing intelligent processing closer to the data sources. This technology opens up great possibilities for IoMT applications, especially when combined with Artificial Intelligence (AI). Edge AI runs AI computations close to the IoT devices and users instead of centralized services such as cloud servers. This paper investigates the potential of Edge AI and IoMT. In this sense, this survey is the first work to further detail Edge AI and Machine Learning Operations in IoMT domains and wearable technology, thus contributing to the literature by comprehensively exploring the potential of ML strategies and operations at the network's edge and intelligence distribution. This study also presents a case study on heart anomaly detection. [ABSTRACT FROM AUTHOR]

Machine Learning as a Service (MLaaS) has been widely used, such as logistic regression models trained in the cloud to provide people with classification predictions. However, these services allow the cloud to collect data from a large number of users, which could raise privacy concerns. With sharing of local models rather than user data, Federated Learning (FL) alleviates data privacy. But, in FL setting the difference between the predictive value provided by the optimal model and the shared global model (called maturing model) that approximates it is not significant, which still raises a security risk. In this paper, we design a flexible and privacy-preserving FL system to tackle these issues. This system makes the training process divide into two stages by analyzing whether the metrics of the iterative global gradient reach a set threshold or not. The two stages are implemented with Re-Encryption and CKKS homomorphism, which not only enables a secure aggregation in the cloud, but also maintains the security of the maturing model. Through detailed analysis, we prove the security of our protocol. Moreover, quantitative comparisons in the final experiments demonstrates that our scheme can achieve trade-off between accuracy, training time cost and communication overhead. [ABSTRACT FROM AUTHOR]

• Collaborating on data collection through multi-mobile sinks reduces the average delay. • Multiple agents acquire optimal strategies through interactions with environment. • Improved experience replay strategy solves the problem of inefficient sampling. • The path planning algorithm increases the task completion rate in complex environment. As one of the most promising research areas in the field of reinforcement learning, multi-agent collaborative decision-making faces the challenge of real-time control in dynamic environments. This paper focuses on the path planning problem of multi-mobile sink collaborative data collection in the Internet of Things. Based on the excellent collaborative decision-making performance of multi-agent deep deterministic policy gradients (MADDPG), a novel path planning algorithm is proposed in the constructed wireless sensor network model. Furthermore, a combined prioritized experience replay strategy is designed to increase the utilization of important experiences in MADDPG. Extensive experiments conducted in various conditions show that compared with traditional MADDPG and MADDPG with prioritized experience replay, the proposed algorithm demonstrates the best performance. The accelerated convergence speed, enhanced training effectiveness and shortened paths indicate the capability of the proposed algorithm in multi-agent collaborative path planning problem. [Display omitted] [ABSTRACT FROM AUTHOR]

This paper proposes an innovative method for achieving optimal tracking control in nonlinear continuous-time systems with input constraints. The method combines reinforcement learning and hybrid metaheuristics to enhance the controller's performance. Specifically, a hybrid metaheuristic algorithm is employed to optimize the hyperparameters of a critic neural network, which serves as the system's controller. The proposed approach is evaluated through extensive simulation studies on a nonlinear system with input constraints. Results demonstrate its superiority over traditional control techniques in terms of accuracy, timeliness, and stability. Notably, the approach effectively eliminates overshoot and steady-state error while providing precise and prompt tracking and showcasing remarkable robustness against model uncertainties. By synergistically integrating reinforcement learning and hybrid metaheuristics, this approach represents a significant advancement in enhancing the control performance of complex nonlinear systems. The simulation studies confirm superiority of the proposed approach over existing techniques, offering a promising solution for achieving optimal tracking control in nonlinear systems with input constraints. This approach holds potential for a wide range of applications, including robotics, aerospace, and manufacturing, where precise and prompt tracking control is critical. [ABSTRACT FROM AUTHOR]

Localization models, particularly for indoor applications like autonomous vehicles and smart manufacturing in Industry 4.0, have emerged as a crucial research area because of the growing need for precise positioning. In this context, distributed cell-free MIMO have demonstrated capabilities for a higher fine-grained resolution and accuracy. This paper explores different approaches to address challenges related to positioning and corresponding data transmission and model compression problems in cell-free massive MIMO systems. We present a novel deep-learning powered CSI compression model for reducing the signalization data associated with the localization task, minimizing the fronthaul capacity requirement by at least a factor of 2, without a significant loss of accuracy. In our research, we also focus on the challenges of deploying deep end-to-end neural networks on resource-constrained embedded platforms. Two approaches are proposed: model modification and magnitude-based pruning. Our results show that the pruned end-to-end localization model has a smaller size by 36% at the cost of an increase on the localization error by a factor of 1.5 (precision is reduced from approximately 5 mm to 7.8 mm). [ABSTRACT FROM AUTHOR]

Many IoT devices are presently in use without sufficient security measures. The vulnerability of these devices to malware highlights the necessity for effective methods to identify malicious activity while protecting data privacy. Federated Learning is a collaborative learning technique that can create a fair model while keeping the local data private. To this end, this paper presents the SIM-FED, a novel model for detecting malware in IoT devices. This model utilizes both deep learning and federated learning algorithms. The SIM-FED offers a privacy-preserving solution for IoT malware detection, without sharing data, enhancing security in distributed environments. The model utilizes a lightweight one-dimensional CNN with optimized hyperparameters, reducing preprocessing, and computational overhead. Various federated aggregation strategies are evaluated, and the FedAvg strategy is selected to integrate the results of local models in the proposed model. Moreover, the model's resilience against white-box and black box cyber-attacks is assessed, revealing minimal performance degradation and highlighting its robustness, which is often overlooked in previous studies. The performance of the SIM-FED is assessed by conducting a series of experiments using the IoT-23 dataset. The results show that the SIM-FED outperforms other deep learning and federated learning models in terms of all evaluation metrics, achieving a remarkable 99.52 % accuracy. [ABSTRACT FROM AUTHOR]

This paper presents an effective and accurate method to analyze the network traffic risk of enterprise public cloud financial system by using deep learning algorithm. To build the analytical model, the collected data is preprocessed to extract relevant features that contribute to risk analysis. The pre-processed data is used to train the deep learning model. The model is designed to learn patterns and relationships in network traffic data. The network traffic risk analysis module is constructed by combining the deep learning model with the system architecture of the enterprise public cloud financial system. Through a trained deep learning model, the module is able to accurately identify abnormal behavior in network traffic and provide corresponding risk assessments. The results show that the proposed system module can accurately identify abnormal behavior in network traffic and provide risk assessment. [ABSTRACT FROM AUTHOR]

Electric vehicles' growing need on in-car and inter-car connection can cause major issues for the infrastructure. By providing a secure and trustworthy intelligent framework, this article will mainly tackle the problem of cyberattacks in electric cars and help to keep them safe from hackers. This paper introduces a novel approach to cyber security analysis that makes use of blockchain technology for smart cloud computing and fuzzy machine learning, both of which are grounded in electric car technology. To monitor and transmit data from electric vehicles, this case employs the smart grid integrated cloud computing model, while to evaluate potentially dangerous actions, the fuzzy adversarial Q-stochastic model (FAQS) is used. Based on kinds of users who have the proper access permissions towards authorized and unauthorized users in accordance with their responsibilities as defined by role-based access control policies, data is encrypted as well as decrypted. A variety of cyber security data sets are subjected to an experimental examination in terms of security rate, RMSE, quality of service, scalability, and energy efficiency. Proposed technique attained energy efficiency 98%, QoS 96%, scalability 91%, security rate 95%. [ABSTRACT FROM AUTHOR]

Because of the openness and complexity of the mobile network environment, the problem of network security is becoming more and more serious, and the mobile network teaching system is also facing the risk of intrusion and attack. Therefore, this paper designs an intrusion detection system for mobile network teaching system. The system uses machine learning and data mining technology to analyze and process the data flow of the mobile network teaching system, extract the characteristics of the data flow, build the intrusion detection model, and realize the intrusion detection of the mobile network teaching system. The research results of this paper show that the intrusion detection system of mobile network teaching system based on intelligent algorithm can effectively detect the intrusion behavior of mobile network teaching system, with high detection accuracy and low false alarm rate, which provides a guarantee for the security and stability of mobile network teaching system. [ABSTRACT FROM AUTHOR]

his paper begins with the assumption that a mobile phone is placed in the right trouser pocket. Its primary objective is to ascertain whether the mobile phone belongs to the driver or the passenger. To achieve this goal, the paper conducts an analysis of raw data collected from phone sensors, specifically the accelerometer, gyroscope, orientation, and GPS, in order to identify and detect movement in the right leg. Based on this analysis, the application makes the decision to disconnect the signal and Internet access on the driver's phone if the car's speed exceeds 60 miles per hour, while maintaining these services for passengers. Subsequently, relevant features are extracted from the data, and a variety of machine learning algorithms are evaluated to determine the most suitable model for the task. The results demonstrate promising performance, suggesting that the proposed method could serve as an effective tool for detecting distracted driving. [ABSTRACT FROM AUTHOR]

Cyber Threat Intelligence (CTI) contains numerous Indicators of Compromise (IOCs) and contextual information, crucial for understanding threat actors' behavior and intentions. However, current information extraction predominantly relies on supervised learning algorithms, presenting challenges in the field of CTI for two reasons. Firstly, the scarcity of labeled data with IOCs hampers the effectiveness of supervised learning. Secondly, existing methods struggle to extract comprehensive contextual features, posing difficulties in IOC recognition within CTI. To address these limitations and better suit the unique characteristics of CTI text, this paper introduces STIOCS, a semi-supervised framework that combines active learning and self-training for IOC extraction. STIOCS enhances IOC extraction accuracy and efficiency by leveraging limited labeled data and a rich unannotated corpus. Firstly, the Active Learning (AL) approach uses the Density-based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to select reliable samples that can reduce noise pollution on pseudo-labeling in self-training. The extraction model integrates Convolutional Neural Network (CNN) and Recurrent Neural Network (RNN) algorithms to extract local and sequential features from CTI text, respectively. Then, the semantic features are enhanced by using the different sizes of convolutional kernels to fuse the two types of features. Finally, the Conditional Random Fields (CRF) layer is employed to recognize IOC entities. Our experimental results demonstrate the effectiveness and robustness of our proposed method in IOC extraction, even with limited labeled data. Compared to supervised methods, our proposed method is only approximately 40% of the dataset is labeled, the F1 scores are achieved better than the existing methods and exhibit consistent performance improvements as the dataset size increases. STIOCS effectively suppresses weak label noise, reduces training costs, and enhances the recognition model's performance. It provides a cost-effective training framework for entity extraction in cyber threat intelligence. [Display omitted] • This paper proposes a semi-supervised active learning framework, STIOCS, aimed at improving the efficiency of IOC extraction in CTI and addressing the challenge of model degradation due to inadequate IOC annotation data. • Firstly, we extract valuable information from unlabeled data during self-training, but pseudo-labeling noise can potentially degrade model accuracy. • Secondly, we propose an active learning method aligned with the training objective to effectively select samples that enhance model performance. • Finally, we conduct extensive experiments on IOC extraction tasks using the fusion model, demonstrating superior performance compared to existing methods. [ABSTRACT FROM AUTHOR]

• A hybrid method is proposed for accurate detection of high impedance fault. • Empirical mode decomposition is used for feature extraction. • EMD with deep learning technique obtained better accuracy in distribution system. • Approximate fault location also identified by proposed technique. An uninterrupted and well-managed energy distribution system requires an effective protection system against various faults and uncertainties. Power system networks have a built-in issue called High Impedance Faults (HIFs), which can't be detected by standard protection mechanisms. This paper deals with the aforementioned issue by combining a deep learning technique (DLT) with a signal decomposition method known as Empirical Mode Decomposition (EMD). This approach utilizes the EMD for the extraction of multiple features from the input current signal, and LSTM neural network is employed for the accurate classification of HIF signals. The comparison results with various EMD methods and various switching events are also performed. Finally, this paper proposes an approximate location identification of HIFs. The simulation results demonstrate the effectiveness of the proposed method for detection in various fault scenarios. [Display omitted] [ABSTRACT FROM AUTHOR]

• Artificial intelligence in penetration testing and vulnerability assessment. • Literature review of penetration testing and vulnerability assessment. • Exploitation and vulnerability analysis against test networks. Vulnerability assessment (e.g., vulnerability identification and exploitation; also referred to as penetration testing) is a relatively mature industry, although attempting to keep pace with the diversity of computing and digital devices that need to be examined is challenging. Hence, there has been ongoing interest in exploring the potential of artificial intelligence to enhance penetration testing and vulnerability identification of systems, as evidenced by the systematic literature review performed in this paper. In this review, we focus only on empirical papers, and based on the findings, we identify a number of potential research challenges and opportunities, such as scalability and the need for real-time identification of exploitable vulnerabilities. [ABSTRACT FROM AUTHOR]

Highlights • Intelligent Hybrid Context-Aware Model for Patients Under-Supervision in Home (IHCAM-PUSH) is proposed to monitor patients with chronic diseases in real-time remotely. • A Hybrid Classification Module (HCM) is proposed to classify patient's health status on dual-mode, on the cloud side (online) or on local side (offline). • The proposed HCM handles the problem of imbalanced datasets in big data. • This paper presents a case study for monitoring elderly patients suffering from blood pressure disorders using (IHCAM-PUSH). Abstract This paper proposes an intelligent hybrid context-aware model for patients under supervision at home that adopts a hybrid architecture with both local and cloud-based components. The cloud-based portion of the model facilitates storing and processing the big data generated by ambient assisted living systems that are used to monitor patients suffering from chronic diseases in their homes, particularly the elderly. The local portion of the model monitors patients in the event of internet disconnections or any other failure in the cloud system. The proposed model utilises context-aware techniques by monitoring physiological signals, ambient conditions, and patient activities simultaneously to derive the real-time health status of the patient. Experimental results demonstrate the effectiveness of our proposed model for monitoring patients and accurately detecting emergencies in imbalanced datasets through a case study on patients suffering from blood-pressure disorders. [ABSTRACT FROM AUTHOR]

Objective In this paper, the proposed model is intended to employ a novel evolutionary computing-based artificial intelligence or machine learning scheme for regression tests to be used for reusability estimation. Such enhancement can lead to accurate reusability pattern estimation, which can be effective for optimal software design purposes. This model is popularly called an aging-resilient software reusability forecast representation. The proposed system employs predominant object-oriented software metrics, such as Chidamber and Kemerer's metrics to examine reusability. Here, cumulative metrics, object-oriented metrics, McCabe's metrics, cohesion and a coupling-based reusability assessment model have been proposed which could be of paramount significance in software design optimization. In this paper, software metrics algorithms and their primary constructions have been developed for estimating the metrics from the UML/class diagrams. It is feasible to derive an efficient and robust reusability prediction model for web-service products using object-oriented metrics. Here, it was also found that OO-CK metrics, particularly complexity, cohesion and coupling-related metrics can be helpful in predicting reusability in web-service software products. Considering the above-mentioned key contributions, it can be stated that the proposed research could be of paramount significance in next-generation software computation systems, primarily for software component reusability, reliability, survivability, aging prediction and stability, and for software excellence assurance purposes. [ABSTRACT FROM AUTHOR]

Scene texts are the type of texts that we see in our surroundings when we are outside. They have custom fonts, sizes, shapes and colors that are often different from the fonts or styles typically found in documents. Scene text recognition has grown in popularity due to its significance in the development of smarter automated systems. Despite the fact that numerous types of research on scene text detection and recognition have been conducted in different languages, similar work on the Bengali language is limited by the lack of large-scale usable datasets. In this paper, we propose a multipurpose system for Bengali Scene text that allows for easy dataset collection via crowdsourcing as well as annotation, classification and detection tasks to be performed on the same platform. It comes with an Android-based application that can be used to capture or load images from the device, select Bengali texts over the images and annotate them. The images and labels are stored in the cloud. A Python-based script is used to perform real-time data processing, analysis, and detection tasks. The proposed system has also been tested on a crowdsourced Bengali scene text dataset collected by this system. The classification model achieved an accuracy score of 97%, and the detection model achieved a Mean Average Precision score of 92 on the on these crowd contributed dataset. [ABSTRACT FROM AUTHOR]

Aiming ultra-reliable and mountable connectivity of tremendously high data rates in the Tb/s era with zero-perceived latency of 6G systems, next generation smart grids will need to captivate the advantages of breakthrough novel technology perceptions, incorporating THz wireless links, broadband and machine learning-based models designs, protocols and management practices. Motivated by the transformation potential of 6G systems, this paper deals with an island detection approach in 6G paradigm for an active distribution network. The proposed island detection method is tailored to adopt the breakthrough technologies of 6G, particularly, ultra-low latency, high volume data transmission and machine learned intelligent system. This methodology calculates the angular sum of the positive and the zero-sequence phase angle of the voltage at point of common coupling combined with random forest (RF) machine learning to identify islanding conditions. The technique has a robust resistance to noise due to the inherent nature of RF, has no consequences on power quality and exhibits a zero non-detection zone. It has the ability to differentiate between islanding and non-islanding events precisely even during active and reactive power discrepancy and transients similar to islanding. The method depicts a high testing accuracy of 98.46% with an island detection time of 52 ms. [ABSTRACT FROM AUTHOR]

• Allows the system to learn from training data and prior information, improving decision-making. • Uses Stochastic Gradient Descent (SGD) optimization to process large datasets in real-time. • Accurately represents indoor environments and improves navigation. • The proposed method efficiently plans collision-free paths for robots and unmanned aerial vehicles indoors. • Potential for real-world implementation. This paper proposes the application of informed machine learning technique to enhance the performance of an autonomous indoor navigation system by leveraging prior knowledge and additional data during training. The system includes simultaneous localization and mapping method for perform two prime functionalities of the localization, and mapping. By integrating machine learning, the system achieves a higher level of autonomy. It involves incorporating extra data alongside a priori knowledge gained from system training, resulting in improved efficiency and autonomy. The optimization of the proposed system model's training process utilizes the stochastic gradient descent algorithm to efficiently handle large volumes of real-time data. Through simulations, the effectiveness of the suggested informed machine learning technique is demonstrated, showcasing its superior performance in autonomous indoor navigation. [Display omitted] [ABSTRACT FROM AUTHOR]

• An effective approach for class incremental learning. • Simulate the human brain's ability to acquire new knowledge while integrating old knowledge. • Isolate the learning environments of new knowledge and old knowledge to mitigate catastrophic forgetting. • It outperforms state-of-the-art methods on avoiding catastrophic forgetting. Humans are capable of acquiring new knowledge on a constant basis, while integrating and optimizing old knowledge without forgetting them. This is mainly attributed to the human brain's ability of partitioned learning and memory replay. In this paper, we simulate this ability and propose an incremental learning network of Sustainable Regional Isolation and Integration (SRII). SRII consists of two phases, regional isolation and regional integration, which are iterated to achieve continuous incremental class learning. Regional isolation isolates new learning processes to avoid interfering with existing knowledge, while regional integration uses knowledge distillation and a margin loss regularization term. The knowledge distillation transfers replay knowledge for alleviating catastrophic forgetting, while the margin loss regularization term is to clarify the boundaries of new and old knowledge for alleviating recency bias. Experimental results on the CIFAR100 and miniImageNet datasets demonstrate that SRII outperforms the state-of-the-arts on avoiding catastrophic forgetting. In all 5-stage and 10-stage incremental settings, SRII outperforms the baseline and achieves at least 5.09%+ average accuracy improvement. Our source code is available at https://github.com/Wuziyi123/SRII. [Display omitted] [ABSTRACT FROM AUTHOR]

Computer generated security concerns have become more modern and complex. Intrusion detection(ID) is a practical issue in the field of computer security whose primary objective is to detect rare attack or assaults and to ensure the security of interior systems. This paper also proposes a semi-class intrusion detection method that combines multiple classifiers to arrange exceptions and typical exercises in a computer system. The abuse detection model is constructed in the light of the decision tree learning-iterative dichotomise 3(DTL-ID3) and is assembled by utilizing the gathered data based on anomaly detection model executed by one class-support vector machine(OC-SVM). In recent years, people have paid more attention to ID/intrusion prevention system (IDS / IPS), which is closely related to the protection and utilization of system management. A few machine-learning standards including neural system, direct hereditary programming, and advanced support vector machines(ASVMs), Bayesian system, multivariate versatile relapse splines, fluffy derivation systems(FIS) and other analogical systems have been researched for the outline of intrusion detection system. In this paper, we build up an amalgam method based on DTL-ID3 and OC-SVM(A-DT and SVM) and evaluate the performance of the projected methodology by using a specific dataset and a crossover method in order to enhance the accuracy of IDS/IPS when contrasted with a singular support vector machine. [ABSTRACT FROM AUTHOR]

A new speech feature extraction method called Mel Modified Group Delay coefficients (MMGDCs) is presented in this paper. In this method, the modified group delay spectrum detects the high formants frequencies, while the Mel filters select these desired formants in the high frequency regions. Also in this paper, a scores fusion approach is proposed between MMGDCs, Mel coefficients (MFCCs) and their extensions using the asymmetric tappers. The adaboost algorithm is used as strategy of this fusion. The performances evaluation of the proposed features and their extended variants are carried out on NIST 2000 corpus, and tested in both clean and simulated noisy conditions, using different noise categories extracted from the NOISEX-92 database. The obtained results show the superiority of the proposed MMGDCs against MFCCs in terms of error reduction, and the power of adaboost algorithm to make the fusion between MMGDCs and MFCCs better, especially under noisy environments. [ABSTRACT FROM AUTHOR]

Multi-layer perceptron neural networks (MLP-NN) are widely utilized in forecasting applications. However, optimal training of these networks is still a challenge. A comprehensive assessment of MLP training approaches comprising of three stages is performed in the present paper. First, the prediction performance is evaluated using twelve training algorithms. Next, optimization algorithms are utilized to enhance the best obtained network parameters obtained from the first step and the performance of eight optimization algorithms is evaluated. Finally, a novel modification is used to improve the performance of the optimization algorithms. The proposed methodologies are applied to two case-studies and statistical metrics are employed for their efficiency evaluation. Wavelet transformation is used to extract the features which will be fed to the MLP-NN as input data. [ABSTRACT FROM AUTHOR]

Smart process repositories have been developed to effectively manage large collections of business process models. Several key features of such repositories require accurate matching of corresponding activities between a pair of process models. However, the existing matching techniques have not yet achieved the desired accuracy, which impedes the effectiveness of process repositories. To that end, this paper proposes a word-embedding-based approach that significantly improves the accuracy of matching. For a comprehensive evaluation of the proposed approach, we have performed experiments using three state-of-the-art word embeddings, two syntactic measures, six semantic measures, and four datasets. The results show that the use of word embeddings outperforms all the syntactic as well as the semantic similarity measures. Moreover, the use of fastText-based embeddings in our proposed technique achieves the highest F1 score, compared to both Word2vec- and GloVe-based embeddings. [ABSTRACT FROM AUTHOR]

Android is one of the most commonly used mobile operating systems; however, its open-source nature and flexibility of usage attract a lot of attention from cybercriminals. In recent years, the rapid increase in malware has become a major cause of concern amongst Android users. The cybercriminals either aim to exploit confidential information from users or try to corrupt their systems by infecting them with malicious code. In order to make Android systems more secure, several malware detection techniques using static, dynamic, and hybrid analysis have been introduced in recent times; however, these techniques are inaccurate and have low efficiency. The paper not only explains how distinctive parallel classifiers can be used for detecting zero-day android malware but also addresses the oncoming highly elusive vulnerabilities. The proposed methodology combines characteristics from various parallel classifiers using expectation maximization to achieve 98.27% accuracy. [ABSTRACT FROM AUTHOR]

• Review of edge network architectures, applications, and trends in IoT-based healthcare solutions. • Study of algorithms based on machine learning (ML) techniques and their use in conjunction with edge devices for high-risk pregnancy monitoring. • Classification mechanism to predict childbirth outcome in pregnancy: Averaged one-dependence estimators. • Performance evaluation using a real dataset from the Maternity School Assis Chateaubriand through a 10-fold cross-validation method. The development of edge computing has allowed the Internet of Things (IoT) to reach higher levels of operational efficiency, knowledge generation, and decision-making through a better relationship between applications and their users. The principal idea is the processing of large amounts of data close to the network edge, where these data are generated. Hence, in this paper, the use of a machine learning technique, known as averaged one-dependence estimators, is proposed for real-time pregnancy data analysis from IoT devices and gateways. This statistical technique is useful for decentralized pre-processing of data and its intermediate storage, reducing the amount of data to be transferred to the cloud and ensuring operability, even in an event of network failure. The results show that this technique presents accurate results with a low computational cost and could be a useful tool to better take advantage of the potential of IoT solutions in healthcare. [ABSTRACT FROM AUTHOR]

One of the most dynamic industries nowadays is the automotive industry. It has dramatically changed from safe, comfortable, and equipped with many entertainment systems to connected autonomous vehicles during the last decade. Although the "Connected Autonomous vehicles" trend dramatically influences the economy, it increases automation and connectivity. Consequently, vehicles can be considered an Internet of Things (IoT) network. Accordingly, the cyberattack surface is more significant, with many vulnerabilities. Considering the fundamental security weakness of automotive onboard networks, besides their superimposing, we firmly believe that a unified security solution is required to resolve the current automotive security problem. This paper proposes an automotive security solution called Generic Automotive Security Suit (GASS). (GASS) is designed to be a practical security solution deployed in the complicated automotive environment. In addition, it defines a standard secure inter-communication between onboard networks. We also introduce the Lightweight Generic Authentication Protocol (LXAP) as the building block of the GASS. We address different security scenarios which face the industry nowadays. The new proposed system does not affect the complexity level of the automotive network. In addition, our simulation results have shown that the LXAP enables security preservation over vehicles' complex networks. [Display omitted] [ABSTRACT FROM AUTHOR]

Low-power consumption is of utmost importance in modern digital systems-on-chip. Approximate computing is a technique used in error-tolerant applications such as multimedia and machine learning to reduce power. This technique creates an appropriate trade-off between performance and accuracy. Adders, as the core of the computing blocks for many digital systems, have a significant impact on their efficiency. This paper attempts to design new approximate full adders with power as the main optimization goal. The proposed circuits are simulated with HSPICE using CMOS and FinFET technologies at 45 nm and 14 nm technology nodes respectively. The simulation results show that on average, the proposed FAs offer 18% and 22% improvements in the dynamic energy and the static power compared to their recent counterparts. Moreover, at the system level, the proposed designs provide sufficient accuracy for real computational applications such as Gaussian filter, discrete cosine transform, and k-means clustering. [ABSTRACT FROM AUTHOR]

Prediction for Intensive Care Unit (ICU) readmission is conducive to assisting doctors in treatment-related decision making and reducing the risk of relapse after discharge. Recently, existing ICU readmission prediction approaches train each sub-task independently, which prevents the models from using complementary information between these sub-tasks. In this paper, we propose correlation enhanced Multi-Task learning with Pearson and RNN-based Neural Ordinary Differential Equations Model (MP-ROM). In order to enhance the learning of general features and avoid the local optima in single-task training, we construct the Shared-Bottom structure of multi-task learning, which enables multiple tasks to share model structure and parameters. Besides, we add the task correlation score calculated by Pearson correlation calculation, enhancing the association between sub-tasks. Experiment results on MIMIC-III dataset show that MP-ROM achieves the highest average precision and demonstrates that task association enhanced can further improve the predictive performance of ICU readmission risk. [ABSTRACT FROM AUTHOR]

The variable network conditions of end-users demand different resolutions, formats, and bitrate versions of videos to be delivered over the network. Fetching each video from the Content Delivery Network (CDN) burdens all network layers. A promising solution is to leverage Mobile Edge Computing (MEC). This paper presents a Machine Learning based caching and transcoding model, which helps release the burden on the backhaul links of the network. The purposed scheme contains a task scheduler and time estimator. The time estimator predicts the job's transcoding time based on the Virtual Machines (VMs) load. The task scheduler maps the transcoding task to different VMs regarding the cost feasibility, Quality of Service (QoS) of the users, and the cost-to-performance ratio of VMs. For this purpose, we prepare a dataset of 500 videos and transcode each video in every lower representation using Amazon Elastic Compute Cloud (EC2). The time estimator is trained on 77% of the video dataset containing more than 80,000 transcoding time data of different videos. The simulation results show that the proposed scheme outperforms its counterparts in terms of cost, average delay perceived by the user, and backhaul burden. [Display omitted] • Analysis of the cost feasibility for edge-transcoding for on-demand videos. • Development of a scheduler that maps the transcoding tasks to different VMs. • Machine-learning-based decision to pull content from CDN or transcode at the edge. [ABSTRACT FROM AUTHOR]

Nowadays Cyber–Physical System (CPS) is implemented in several areas, such as health care, agriculture, transportation, etc. The main aim of CPS is for automation or minimization of human interaction. The conventional Machine Learning (ML) based approach involves sensitive issues such as information leakage. That issue is resolved using the Federated Learning (FL) based approach partially. The FL uses a decentralized environment for training the models, where models are trained locally, and only the updated part is sent to the data center or cloud server. It improves data privacy as the clients send only the local updates (model gradients). But also there is a high risk of data leakage while publishing the data in FL based System. A Differential Privacy Preservation Technique (DPPT) is an efficient technique to preserve information leakage. In this paper, a differential privacy preservation technique has been proposed by adding random noise to a data sample in order to generate anonymity. The performance of the model, and data quality are evaluated as privacy preservation techniques often degrade data quality. [Display omitted] • The proposed work is based on differential privacy in the Federated Learning environment. • It generates random noise, which is added to actual data to generate anonymity and mitigate adversarial attacks. • The proposed model ensures data privacy and preserves information leakage in a cyber–physical system. • The performance and data quality of the model is measured and found promising results. [ABSTRACT FROM AUTHOR]

In this paper, we consider the optimal control problem of heavy material handling manipulators for agricultural robots. Unlike the existing results on agricultural robots, the robot parameters may be unknown for the designer in this paper. To learn the linear quadratic control gain under unknown robot parameters, two reinforcement learning algorithms, i.e., policy iteration (PI) algorithm and value iteration (VI) algorithm, are proposed. Then, through combining the advantages of PI algorithm and VI algorithm, i.e., satisfactory convergence rate and without the restriction on feasibility initial control policy, respectively, a hybrid iteration (HI) algorithm is proposed, which can both achieve a satisfactory convergence rate and remove restrictions on feasibility initial control policy. It is shown that the convergence of the proposed HI algorithm can be achieved in theory. Finally, a simulation example is given to show that our designed HI algorithm can achieve a satisfactory simulation time. [ABSTRACT FROM AUTHOR]

• The Intelligent Microbat Routing (IMR) and deep learning (DL) based Popularity Content Caching (PCC) IMR-PCC hybrid intelligent architecture for the caching and routing achieved the better QoS performances in vehicular edge networks. • This paper introduced the popularity based content features for achieving the better learning rate in the intelligent caching. • The framework utilized OMNET++ interfaced with Python3.8 platform for the implementation which results in better implementation of the model. • The paper focusses on the joint routing and caching technique for the vehicular edge and IoT networks. • IMR-PCC achieved the better Quality of Service (QoS) as well as executes cache performance which is highly suitable for the vehicular IoT networks. Vehicular networks are considered as one of the paramount components of intelligent transportation systems. With the help of new advanced technologies like Internet of things (IoT), and many wireless applications such as 5 G/6 G, these vehicular networks has changed its face of applications in terms of computations, communications and privacy. The arising vehicular applications such as crash awareness and emergency brake requires better communication, storage resources, and computation with severe execution prerequisites on response time and network bandwidth. But still, efficient data computation with an intelligent routing mechanism still remains to be darker side of the research. To achieve the improvisation in above parameters, this paper proposes the hybrid combination of the Intelligent Microbat Routing (IMR) and Deep learning based Popularity Content Caching (PCC) techniques which has been targeted for IoT based vehicular networks. This paper introduces the Reinforced Gated Recurrent Units (R-GRU) with High Speed Extreme Learning Machines for better prediction of contents' popularity that aids for better caching. The proposed IMR-PCC architecture consists of four different tasks such as popularity prediction, cache placement, data retrieval and effective routing. The extensive experimentation is carried out to evaluate the IMR-PCC by implementing the dynamic network topology with multi factors, which directly impacts on the performance of the complete system. Nearly 300,000 contents were collected randomly and the complete algorithm was implemented in the OMNET++ interfaced with Python 3.8. Finally from the experimental results, the proposed IMR-PCC architecture proved its excellence in terms of accuracy (98.6%), precision (98.6%), recall (98.6%), F1-Score (98.8%), cache hit ratios (0.25), latency (0.3), Packet Delivery Ratio (PDR) (99%) than other existing methodologies such as Q-learning, BLSTME, MPCF, ICN. So the proposed IMR-PCC achieved the better Quality of Service (QoS) as well as executes cache performance which is highly suitable for the vehicular IoT networks. [Display omitted] [ABSTRACT FROM AUTHOR]

Intelligent Transport Systems (ITS) provide a promising technology to enhance road safety. The vehicular standard wireless access in vehicular environment (WAVE), also known as dedicated short-range communication (DSRC), can assist in reducing the number of deadly crashes. However, DSRC has a limited range. To enhance the network coverage, roadside units (RSUs) are placed along the road. However, the placement of RSUs at every instant increases the infrastructure cost. In this paper, we proposed the cooperative vehicular architecture, with network function virtualization (NFV) enabled RSU inside the mobile edge computing (MEC) unit. RSUs are only placed in the dense traffic region. We applied the Long short-term memory (LSTM) based machine-learning algorithm to predict the traffic flow based on the vehicle information table (VIT) maintained at the MEC unit. NFV is implemented at the top of RSU. Based on predicted traffic density it assists RSU to enhance its coverage range by exploiting the transmit power. Furthermore, MEC is also responsible for cooperative relay-based communication. Optimal stopping theory is modeled to select the best candidate relay node immediately. In this paper, we tested the proposed scheme in actual on-road vehicles and through simulations performed in network simulator NS-3. [Display omitted] • The proposed architecture consists of NFV enabled RSU inside MEC. • The proposed scheme is tested on-road vehicles for selecting the optimal relay candidate. • A stacked-layer long short-term memory (LSTM) model is proposed and applied at the MEC. • The NFV layer is devised at the RSU which supports agile services. • The optimal stopping theory is modeled for the selection of relay. [ABSTRACT FROM AUTHOR]

In recent years, cloud computing has emerged as a promising paradigm providing various resources in the Cloud Data Center (CDC), including computations, storage, and even platform as a service. The virtualization technology plays a pivotal role in CDC resources management and provides an enchanting feature of virtual machine (VM) migration which provides several benefits in terms of VM scheduling, fault tolerance, load balancing, energy efficiency, power management, and security. To achieve efficient VM migration, a plethora of VM migration schemes have been proposed in the literature aiming to serve the quality of service-driven user requirements. This paper surveys the most recent and state-of-the-art Artificial Intelligence (AI) based and conventional load balancing, energy-aware, SLA aware, and network-aware live VM migration schemes. Through an extensive literature review, this paper investigates the most critical aspects of conventional and AI-driven live VM migration schemes. Finally, a few open research challenges that require further consideration from the research community are highlighted. [Display omitted] • Overview of cloud computing, data centers, virtualization, and VM migration. • Conventional and AI-based Live VM migration schemes categorization. • Live VM migration survey of load, energy, SLA, and network schemes. • VM migration analysis based on migration cost and AI benefits and drawbacks. • Live VM migration research challenges and future directions. [ABSTRACT FROM AUTHOR]

Modern businesses benefit significantly from advances in computer vision technology, one of the important sectors of artificially intelligent and computer science research. Advanced computer vision issues like image processing, object recognition, and biometric authentication can benefit from using deep learning methods. As smart devices and facilities advance rapidly, current networks such as 4 G and the forthcoming 5 G networks may not adapt to the rapidly increasing demand. Classification of images, object classification, and facial recognition software are some of the most difficult computer vision problems that can be solved using deep learning methods. As a new paradigm for 6Core network design and analysis, artificial intelligence (AI) has recently been used. Therefore, in this paper, the 6 G wireless network is used along with Deep Learning to solve the above challenges by introducing a new methodology named Optimizing Computer Vision with AI-enabled technology (OCV-AI). This research uses deep learning – efficiency algorithms (DL-EA) for computer vision to address the issues mentioned and improve the system's outcome. Therefore, deep learning 6 G proposed frameworks (Dl-6 G) are suggested in this paper to recognize pattern recognition and intelligent management systems and provide driven methodology planned to be provisioned automatically. For Advanced analytics wise, 6 G networks can summarize the significant areas for future research and potential solutions, including image enhancement, machine vision, and access control. [ABSTRACT FROM AUTHOR]

'Fake news' refers to the misinformation presented about issues or events, such as COVID-19. Meanwhile, social media giants claimed to take COVID-19 related misinformation seriously, however, they have been ineffectual. This research uses Information Fusion to obtain real news data from News Broadcasting, Health, and Government websites, while fake news data are collected from social media sites. 39 features were created from multimedia texts and used to detect fake news regarding COVID-19 using state-of-the-art deep learning models. Our model's fake news feature extraction improved accuracy from 59.20% to 86.12%. Overall high precision is 85% using the Recurrent Neural Network (RNN) model; our best recall and F1-Measure for fake news were 83% using the Gated Recurrent Units (GRU) model. Similarly, precision, recall, and F1-Measure for real news are 88%, 90%, and 88% using the GRU, RNN, and Long short-term memory (LSTM) model, respectively. Our model outperformed standard machine learning algorithms. [Display omitted] • The research uses the process of Information Fusion to obtain real news data. • We have used Deep Learning classifiers to propose 39 novel features for text. • The paper proposes sentiment features, linguistic features and name entity base features. After extracting features from the text, our new features detect COVID-19 related fake news with an accuracy of 86.12%. Thus, accuracy is increased by 20% with the novel features. • This paper compares Deep Learning algorithm results with the standard Machine Learning (ML) algorithms using the same fusion of dataset and shows a higher accuracy. Therefore, it confirms the superiority of our Deep Learning model. [ABSTRACT FROM AUTHOR]

• We proposed a 9-layer deep convolutional neural network for leaf disease identification. • We used six different types of image augmentation methods, and compared their validation performances using data augmentation and not using data augmentation. • Fine-tuning the batch size, dropout and epoch improves the performance of our model. • We validated that the proposed model performs better than transfer learning techniques. • Our model produced an average testing accuracy of 96.46%, which was better than the state-of-the-art machine learning approaches. In this paper, we proposed a novel plant leaf disease identification model based on a deep convolutional neural network (Deep CNN). The Deep CNN model is trained using an open dataset with 39 different classes of plant leaves and background images. Six types of data augmentation methods were used: image flipping, gamma correction, noise injection, principal component analysis (PCA) colour augmentation, rotation, and scaling. We observed that using data augmentation can increase the performance of the model. The proposed model was trained using different training epochs, batch sizes and dropouts. Compared with popular transfer learning approaches, the proposed model achieves better performance when using the validation data. After an extensive simulation, the proposed model achieves 96.46% classification accuracy. This accuracy of the proposed work is greater than the accuracy of traditional machine learning approaches. The proposed model is also tested with respect to its consistency and reliability. [ABSTRACT FROM AUTHOR]

Energy prediction is in high importance for smart homes and smart cities, since it helps reduce power consumption and provides better energy and cost savings. Many algorithms have been used for predicting energy consumption using data collected from Internet of Things (IoT) devices and wireless sensors. In this paper, we propose a system based on Multilayer Perceptron (MLP) to predict energy consumption of a building using collected information (e.g., light energy, day of the week, humidity, temperature, etc.) from a Wireless Sensor Network (WSN). We compare our system against four other classification algorithms, namely: Linear Regression (LR), Support Vector Machine (SVM), Gradient Boosting Machine (GBM) and Random Forest (RF). We achieve state-of-the-art results with 64% of the coefficient of Determination R 2, 59.84% Root Mean Square Error (RMSE), 27.28% Mean Absolute Error (MAE) and 27.09% Mean Absolute Percentage Error (MAPE) in the testing set when using weather and temporal data. [ABSTRACT FROM AUTHOR]

Emergence of social media invokes social actors to share their information digitally. Moreover, it enables them to maintain their links with other people worldwide. Due to its massive popularity, it has become a fascinating test bed to initiate various attacks. Attackers form Sybil nodes to disseminate malicious content in order to infect legitimate users aiming to steal sensitive information, such as user's credentials. Therefore, the focus of this paper is to detect malicious profiles, specifically on the Twitter microblogging platform. We propose a hybrid approach which leverages the capabilities of machine learning techniques to identify malicious profiles. Initially, Petri net structure analyzes the user's profile and various features, and these features are then used to train classifier. This is achieved to identify malicious profiles from legitimate users. Finally, to prove the efficiency, a comparative analysis of our approach is conducted with existing state-of-the-art techniques. The experimental results reveal that our approach achieves a high detection rate of 99.16% which is better than other techniques. [ABSTRACT FROM AUTHOR]

Abstract In this paper, we present a deep learning-based recognition algorithm to identify pulsars by observing data containing millions of candidates including radio frequency interference and noise sources. The dataset is obtained from the High Time Resolution Universe survey created and updated by the Parkes telescope. We investigate several effective single and combined features via simple logistic regression. To deal with the imbalanced dataset, we oversimplify the original dataset at different sampling rates, which is also one of the learning parameters. After training the pre-processed dataset via a convolutional neural network, we provide a cross-validated evaluation of all candidates. Results show that the deep-learning based recognition algorithm can identify the pulsar and radio frequency interference signals with high accuracy. The precision and recall of radio frequency interference are both 100%, and those of pulsars are 91% and 94%, respectively. [ABSTRACT FROM AUTHOR]

Abstract This paper presents a new approach to automatically classify hemorrhagic and ischemic strokes using the Structural Co-Occurrence Matrix (SCM) extracted from the main frequencies of computed tomography images of the brain. The main advantage of this approach is that it only uses the image as a parameter. Specificity, sensitivity, positive predictive value, F-Score, harmonic mean, and accuracy were used as metrics to evaluate the efficiency of the SCM. The SCM was compared with local binary patterns, gray-level co-occurrence matrices, invariant moments of Hu and with the feature extraction method based on human tissue density patterns named Analysis of Human Tissue Densities. Multiple machine learning classifiers were used including support vector machine, multilayer perceptron, minimal learning machine and the linear discriminant analysis. The results show that the SCM in the frequency domain can extract the most discriminant structural information of strokes automatically, obtaining good results without the needed of additional parameters. [ABSTRACT FROM AUTHOR]

Abstract Convolutional neural networks (CNNs) have been a focus area of machine learning in recent years, and they are widely used in vision and speech processing because of their superior performance. However, CNNs are usually resource-heavy to ensure higher accuracy, i.e., an accurate network with millions of parameters requires high performance computing devices. This prevents the use of CNNs in resource-limited hardware. In this paper, we propose a novel CNN simplification method to prune feature maps with relatively low discriminability magnitudes, which can produce a simplified CNN with reduced computational cost. Specifically, we define the critical points among the discriminability values of feature maps in each convolutional layer, and use these critical points to easily find the best pruning number of feature maps. Our experimental results show that in each convolutional layer of the VGG model, 15.6% to 59.7% of feature maps can be pruned without any loss of accuracy in classification tasks. [ABSTRACT FROM AUTHOR]

Collaborative filtering based Recommender System is one of the most common technique used for personalized product ranking. It aids the consumer in decision-making process. It helps to choose a product according to the consumer's preference from a large pool of choices.Despite its success, collaborative filtering suffers from the sparsity problem which limits the quality of recommendations. In this paper, we investigate the application of clustering collaborative framework. A unique centroid selection approach for k -means clustering algorithm is proposed that aims to improve clustering quality. The results on three benchmark datasets depict the improvement in the quality of recommendations made. [ABSTRACT FROM AUTHOR]

With the steady growth in the adoption of IoT in an industrial context, security challenges intensify. Security threats to IoT devices in a nuclear plant or an oil refinery are much higher risk compared to home appliances. In this paper, we present an explainable, efficient, and highly accurate machine-learning based intrusion detection system for industrial IoT, referred to as E2I3DS. This work successfully reduces the needed features within the WUSTL-IIOT-2021 dataset from 48 features to 11 only, while maintaining very high accuracy. Experiments utilizing the 11-feature dataset showed that proposed system has a higher accuracy of 99.97%, with a false positive rate of 0.01% and a false negative rate of 0.04%. The proposed system also proved to be very efficient with an excellent detection time of 0.1517 μ s. In addition, the model was explained using Shapley additive explanations. [Display omitted] • Introducing a high-efficiency, and high-accuracy intrusion detection system for Industrial IoT. • Selecting 11 network-flow features as the most effective ones in intrusion detection. • Explaining the impact of the features on the prediction decisions using Shapley Additive Explanation. This ensures that the detection decisions originate from an explainable system rather than a black-box. [ABSTRACT FROM AUTHOR]

Panicle segmentation is a critical step in crop phenotyping research to investigate physiological and structural traits of crops. Unmanned aerial vehicle-based image acquisition integrated with deep learning algorithms replaces labour-intensive field investigation and crop monitoring. The drone has limited resources, so it requires a small but efficient deep learning model. This paper proposes a novel Tiny Criss-Cross Network (TinyCCNET) deep learning model, based on criss-cross attention and a tiny approach. The criss-cross attention method captures contextual information from all pixels, enhancing segmentation accuracy, while the tiny approach is applied to the backbone network ResNet50. The proposed model outperforms the other existing models with 86.5% accuracy, 81.6% mIoU, and 12.3 GFLOPS. TinyCCNET achieves high accuracy, while being resource-efficient, making it an ideal choice for agricultural drone applications. [Display omitted] • Deep learning models applied to UAV requires a tiny model with higher performance. • TinyCCNET model based on cris cross attention and the tiny approach is proposed. • It is implemented on paddy rice imagery dataset, which consists of 8400 images. • It achieves 86.5% accuracy, 81.6% mIoU with 12.3 GFLOPS. [ABSTRACT FROM AUTHOR]

The Internet of Things (IoT) platform allows physical devices to connect directly to the internet and upload data continuously. Insecure access makes IoT platforms vulnerable to different network intrusion attacks. As a result, the Intrusion Detection System (IDS) is a core component of a modern IoT platform. However, traditional IDS often follows rule-based detection where the rules can be changed and exposed to the attacker and becomes weak over time. An efficient IDS also needs to be dynamic and effective in real time. This paper proposes a deep learning-based algorithm to protect the network against Distributed Denial-of-Service (DDoS) attacks, insecure data flow, and similar network intrusions. A system architecture is designed for a cloud-based IoT framework to implement the proposed algorithm efficiently. The performance evaluation using standard datasets demonstrates that the proposed model provides an accuracy of up to 99.99%. [ABSTRACT FROM AUTHOR]

Many of the proposed machine learning (ML) based network intrusion detection systems (NIDSs) achieve near perfect detection performance when evaluated on synthetic benchmark datasets. However, there is no record of if and how these results generalise to other network environments. In this paper, we investigate the cross-domain performance of ML-based NIDSs by extensively evaluating eight supervised and unsupervised learning models on four recently published benchmark NIDS datasets. Our investigation indicates that none of the considered models is able to generalise over all studied datasets. Interestingly, our results also indicate that the cross-domain performance has a high degree of asymmetry, i.e., swapping the source and target domains can significantly change the classification performance. Our investigation also indicates that overall, unsupervised learning methods perform better than supervised learning models in our considered scenarios. We further used SHAP values to explain the observed cross-domain performance results. They show a high correlation between a good model performance and a correspondence between feature distributions/values and Attack/Benign classes. [ABSTRACT FROM AUTHOR]

This paper focuses on integrating two popular signal processing techniques, i.e., variational mode decomposition (VMD) and discrete wavelet transform (DWT), with deep learning (DL) and machine learning (ML) algorithms to come up with nine novel ensemble models for solar radiation forecasting. For these ensemble models, we considered five DL algorithms that include gated recurrent units (GRU), long short-term memory (LSTM), bidirectional long short-term memory (BiLSTM), convolutional neural network (CNN), and deep neural network (DNN); two ML algorithms that include artificial neural network (ANN) and support vector regression (SVR). The proposed nine models were tested for seven Indian cities, namely Delhi, Chennai, Hyderabad, Nagpur, Patna, Trivandrum, and Bhubaneshwar. We used root mean square error (RMSE), mean absolute error (MAE), and coefficient-of-determination (R2) metrics for understanding the performance. Observed that the DL-based VMD integration had generated considerably promising results compared to DWT integration. Out of the nine models, VMD-integrated GRU gave the most optimum results for all the cities with RMSE (0.82-1.22), MAE (0.54-1.02), and R2 (0.83-0.93). It's because GRU employs fewer training parameters, which requires less memory and functions faster than any other algorithm. Overall, this study helps Indian cities to forecast solar radiation more effectively with advanced ML and DL algorithms and scope for applying these models elsewhere around the world. [ABSTRACT FROM AUTHOR]

• Disease prediction based on different user symptoms is done. • Different machine learning-based approaches are designed and employed for different tasks. • Utilizes sensors to measure corresponding symptoms, the study combines the IoT and cutting-edge deep learning models to yield an optimal result. • Microservice-based user interface is designed. COVID-19 disrupted lives and livelihoods and affected various sectors of the economy. One such domain was the already overburdened healthcare sector, which faced fresh challenges as the number of patients rose exponentially and became difficult to deal with. In such a scenario, telemedicine, teleconsultation, and virtual consultation became increasingly common to comply with social distancing norms. To overcome this pressing need of increasing 'remote' consultations in the 'post-COVID' era, the Internet of Things (IoT) has the potential to play a pivotal role, and this present paper attempts to develop a novel system that implements the most efficient machine learning (ML) algorithm and takes input from the patients such as symptoms, audio recordings, available medical reports, and other histories of illnesses to accurately and holistically predict the disease that the patients are suffering from. A few of the symptoms, such as fever and low blood oxygen, can also be measured via sensors using Arduino and ESP8266. It then provides for the appropriate diagnosis and treatment of the disease based on its constantly updated database, which can be developed as an application-based or website-based platform. [Display omitted] [ABSTRACT FROM AUTHOR]