Abstract Autonomous intelligent systems are artificial intelligence (AI) tools that act autonomously without direct human supervision. Cloud computing (CC) and Internet of Things (IoT) technologies find it challenging to deploy sufficient security defences because of the different structures, storage, and limited computing capabilities that make them more vulnerable to attacks. Security threats against IoT structures, devices, and applications are increasing with the demand for IoT technology. The training data available to AI models may be limited, which could impact their performance and generalizability. Adopting AI solutions in real‐world situations may be impeded by compatibility concerns and the requirement for flawless integration. Malware classification errors can occur due to a lack of contextual knowledge, particularly in cases where benign files behave identically to malicious. Various studies were carried out on detecting IoT malware to evade the menaces posed by malicious code. However, prevailing techniques of IoT malware classification supported particular platforms or demanded complicated methods for attaining higher accuracy. This study introduces an enhanced slime mould optimization with a convolutional BLSTM autoencoder‐based malware classification (ESMO‐CBLSTMAE) system in the IoT cloud platform. The projected ESMO‐CBLSTMAE system focuses on detecting and classifying malware in the IoT cloud platform. To achieve that, the ESMO‐CBLSTMAE algorithm employs a min–max normalization technique for scaling the input dataset. The ESMO‐CBLSTMAE method uses a convolutional bidirectional long short‐term memory autoencoder (CBLSTM‐AE) model for the malware detection process. Lastly, the ESMO method is executed for the optimum hyperparameter tuning of the CBLSTM‐AE technique, which boosts the malware classification results. The experimental analysis of the ESMO‐CBLSTMAE method is tested against a benchmark database, and the outcomes portray the greater efficacy of the ESMO‐CBLSTMAE approach over other existing techniques. The proposed malware classification model achieved an accuracy of 98.57 and F Score of 80.77 and outperformed the existing models.

Abstract In the realm of Drug‐Target Interaction (DTI) prediction, this research investigates and contrasts the efficacy of diverse evolutionary algorithms in fine‐tuning a sophisticated hybrid deep learning model. Recognizing the critical role of DTI in drug discovery and repositioning, we tackle the challenges of binary classification by reframing the problem as a regression task. Our focus lies on the Convolution Self‐Attention Network with Attention‐based bidirectional Long Short‐Term Memory Network (CSAN‐BiLSTM‐Att), a hybrid model combining convolutional neural network (CNN) blocks, self‐attention mechanisms, and bidirectional LSTM layers. To optimize this complex model, we employ Differential Evolution (DE), Particle Swarm Optimization (PSO), Memetic Particle Swarm Optimization Algorithm (MPSOA), Fire Hawk Optimization (FHO), and Artificial Hummingbird Algorithm (AHA). Through thorough comparative analysis, we evaluate the performance of these evolutionary algorithms in enhancing the CSAN‐BiLSTM‐Att model's effectiveness. By examining the strengths and weaknesses of each algorithm, our study aims to provide valuable insights into DTI prediction, identifying the most effective evolutionary algorithm for hyperparameter tuning in advanced deep learning models. Notably, Fire‐hawk optimization (FHO) emerges as particularly promising, achieving the highest Concordance Index (C‐index) as 0.974 for KIBA datasets and 0.894 for DAVIS datasets and demonstrating exceptional accuracy in ranking continuous predictions across both the datasets.

ABSTRACT The quality of human‐computer interactions (HCI) has increased recently because of developments in artificial intelligence (AI) and machine learning methods, but there are still numerous obstacles to overcome. One of these difficulties that has been taken into account by several academics in recent years is the recognition of emotions via the processing of facial pictures. Most of the previously suggested solutions have drawbacks like poor accuracy and restrictions on the amount of emotions detected. On the other hand, researchers need to focus more on identifying the ideal feature set that results in maximum detection accuracy. This work addresses these issues by outlining a novel method for extracting the best face characteristics and their improved categorisation. Pre‐processing, feature extraction, feature selection and classification are the four phases of the suggested technique. Image normalisation and face recognition are steps in the pre‐processing stage. The ideal features are chosen using a black hole optimisation approach in the proposed method, which combines a Convolutional Neural Network (CNN) and Multi‐scale Local Binary Patterns (MLBP) to extract the feature. The next step is to categorise certain characteristics and identify facial emotions in the photos using Error Correcting Output Codes (ECOC). To lessen the issue's complexity, the suggested ECOC model combines a number of Support Vector Machine (SVM) classifiers. Results reveal that the proposed model has average accuracies of 98.9% and 79.82%, respectively, for the Yale and FER‐2013 datasets in recognising facial expressions, which shows an increase of at least 1% over the prior approaches.

Abstract Deep learning and metaheuristic algorithms have recently increased in various sciences, including financial accounting information systems (FAISs). However, the existence of large datasets has dramatically increased the complexity of these hybrid networks, so to address this shortcoming, this paper aims to develop a quantum‐behaved chimp optimization algorithm (QCHOA) and deep neural network (DNN) for the prediction of the profit based on FAISs. Considering that there is no suitable dataset for the challenge, a novel dataset is developed utilizing the 15 features from the Chinese market dataset to compare more. This work designs QCHOA and five DNN‐based predictors to forecast profit. These algorithms include the universal learning CHOA (ULCHOA), the niching CHOA (NCHOA) as the two best‐modified versions of CHOA, the quantum‐behaved whale optimization algorithm (QWOA), and the quantum‐behaved grey wolf optimizer (QGWO) as the two best quantum‐behaved optimizers as well as classic CHOA. The most effective deep learning‐based predictors for forecasting the profit, ranked from highest to lowest, are DNN‐QCHOA, DNN‐NCHOA, DNN‐QWOA, DNN‐QGWO, DNN‐ULCHOA, DNN‐CHOA, and classic DNN, with corresponding ranking scores of 42, 36, 30, 24, 18, 12, and 6. As a final suggestion for profit prediction, the DNN‐CHOA is shown to be the most accurate model.

Smart logistics industry leverages advanced software and hardware systems to enable efficient transmission. The incorporation of smart technologies, including digital twin (DT) and blockchain assesses vulnerabilities in the logistics industry, making them effective for physical attacks by users for stealing and theft control. DT persists a transformative potential in optimizing industrial operations. By bridging the physical and digital worlds, they enable real-time monitoring, predictive analytics, and enhanced decision making, driving innovations in efficiency, security, and sustainability. Conspicuously, the primary objective is to propose an effective logistic monitoring system for ensuring automated theft control. Specifically, the proposed model determines the logistic transmission patterns through secure surveillance using Internet of Things-empowered blockchain technology. Moreover, the deep learning technique of a bi-directional convolutional neural network is used to assess theft and stealing vulnerability by users in real-time for optimal decision making. The proposed approach has been demonstrated to enable accurate real-time analysis of vulnerable behavior. Based on the experimental simulations, the suggested solution effectively facilitates the development of superior logistic monitoring. The performance of the proposed system is evaluated using several statistical metrics, including latency rate (26.15 s), data processing cost, prediction efficiency (accuracy (96.12%), specificity (97.53%), and F-measure (97.25%), reliability (93.34%), and stability (0.74).

National security is one of the premier sectors of research and development in every country. Moreover, government organizations spend large funds to ensure technical advancements in the defense sector. However, major flaws in the routine activities of security personnel have resulted in severe tragedies. Conspicuously, a comprehensive Internet of Things (IoT)-based methodology has been presented to evaluate the integrity of an officer based on professional and personal activities. Specifically, the current study proposes a digital twin-inspired method for evaluating the overall performance of intelligence agency officers (IAOs). The probabilistic digital integrity estimate (DIE) is formalized through data analysis using the Bayesian belief model (BBM). Additionally, the multiscaled long term memory (MLSTM) model has been proposed to predict the overall integral behavior of IAO. The proposed system has been verified over a real-world data set with 42892 instances. Results show that the proposed technique surpasses comparative models in key metrics, such as temporal delay effectiveness (127.79 s), categorization analysis [precision (96.67%), specificity (97.08%), and sensitivity (97.55%)], decision-modeling efficacy [specificity (93.49%), precision (93.49%), and sensitivity (93.69%)], reliability (94.86%), and stability (82.0%).

Medical professionals have devised novel solutions to transform the healthcare industry. Modern technology of digital twins (DTs) can revolutionize medical treatment significantly. The DT technology incorporates digitizing physical entities by constantly monitoring their current status. Conspicuously, a state-of-the-art secure framework for monitoring adults’ physical activity is formulated using the culmination of the DT technology with Internet of Things (IoT)-edge computing, and blockchain technology. The presented framework is designed to discreetly secure the health data of the individual. To identify healthcare vulnerabilities in adults, the present study employs deep learning’s ability to analyze IoT data sequentially. Specifically, a deep learning-assisted multilayered convolutional neural networks (CNNs) and long short-term memory (LSTM) technique is proposed for real-time vulnerability assessment. Additionally, the proposed framework can protect personal healthcare data by using the blockchain technique. For performance validation, numerous simulations were performed over the challenging data set. Based on the results, the proposed methodology can outperform state-of-the-art techniques by registering enhanced values of Temporal Delay Efficacy (120.79 s), Prediction Efficacy (Accuracy (92.24%), Specificity (94.67%), Sensitivity (95.26%), and F-measure (95.69%)), Reliability (91.58%), and Stability (64%).

Over the last decade, Dengue infection has expanded more rapidly than any other viral illness. The current research investigates the vast potential of the Internet of Things (IoT), and Edge–cloud computing in reproving dengue virus (DGN) infection-related technological healthcare solutions. Specifically, a hierarchical healthcare framework is proposed for preventing the spread of DGN using Edge–cloud-assisted IoT technology. The presented system can monitor and forecast an individual’s susceptibility to DGN infection in a ubiquitous manner. Using K-means clustering, the presented system determines an individual’s DGN infection status and generates alert signals in real-time. In addition, the proposed technique employs cloud computing to monitor people healthcare impacted by DGN. Moreover, it ensures probabilistic predictions about susceptibility to the DGN virus using Bayesian belief networks and artificial neural networks. The proposed system can assess health vulnerability, thereby reducing the probability of health loss. The suggested system’s validity and applicability are confirmed by experimental evaluation. The simulation results of the proposed system confirm its optimal performance in terms of temporal delay (14.15 s), classification efficacy [accuracy (91.66%), sensitivity (92.34%), specificity (90.25%), and F-measure (91.12%)], prediction effectiveness [error (0.23), and pearson coefficient (85%)], and stability (72%).

The significance of intelligent transportation is increasing in modern societies. The development of electric mobility is a result of extensive research and industrial needs. Conspicuously, the current study proposes a smart Electric Vehicular (EV) performance system for the transportation industry that uses IoT-Fog-Cloud (IFC) computing technology to provide an effective analysis of domestic and commercial EVs. The system analyzes real-time EV-oriented attributes to present a Performance Analysis Measure (PAM). The framework uses a Bayesian Belief Model (BBM) to classify EV-related attributes in different categories over a temporal scale. Finally, a two-level threshold-based decision tree model is proposed for an overall assessment of the EV. Experimental simulations were performed to validate its effectiveness over challenging datasets with nearly 56365 data instances. Comparative to state-of-the-art techniques, the proposed framework registered enhanced performance for statistical metrics of Delay Assessment (126.68s), Statistical Classification Analysis (Specificity (96.97%), Precision (95.56%), and Sensitivity (96.44%)), Decision-making efficiency (97.53%), Reliability (92.69%), and Stability (0.73).

Internet of Things (IoT) technology has revolutionized the industrial sector. This research article focuses on the development of Food Industry 4.0, which was made possible by advancements in edge-cloud computing and IoT technologies. The study presents an IoT-based smart framework that uses the Bayesian belief network (BBN) on the edge-cloud platform to analyze data in the food industry. The acquired data is assessed to estimate the probability of food quality (PFQ) and evaluate food outlets using the food quality analysis measure (FQAM). Additionally, a Bi-level decision-tree modeling is presented to assess food quality. Food-oriented data security is ensured using blockchain. The proposed model is tested on a complex data set containing data about four restaurants with about 43 520 individual instances. Simulations show effective results of temporal delay (94.41 s), decision-making efficacy (99.64%), classification efficiency (precision (96.67%), specificity (96.97%), and sensitivity (97.55%)), stability (74.25%), and reliability (93.70%).

Twig pattern matching is a core operation in XML query processing because it is how all the occurrences \nof a twig pattern in an XML document are found. In the past decade, many algorithms have been proposed \nto perform twig pattern matching. They rely on labelling schemes to determine relationships between \nelements corresponding to query nodes in constant time. In this paper, a new algorithm TwigStackPrime is \nproposed, which is an improvement to TwigStack (Bruno et al., 2002). To reduce the memory consumption and \ncomputation overhead of twig pattern matching algorithms when Parent-Child (P-C) edges are involved, TwigStackPrime \nefficiently filters out a tremendous number of irrelevant elements by introducing a new labelling \nscheme, called Child Prime Label (CPL). Extensive performance studies on various real-world and artificial \ndatasets were conducted to demonstrate the significant improvement of CPL over the previous indexing and \nquerying techniques. The experimental results show that the new technique has a superior performance to the \nprevious approaches.