Computers, Volume 12, Issue 3 (March 2023) – 22 articles

By using computer-aided arrhythmia diagnosis tools, electrocardiogram (ECG) signal plays a vital role in lowering the fatality rate associated with cardiovascular diseases (CVDs) and providing information about the patient’s cardiac health to the specialist. Current advancements in deep-learning-based multivariate time series data analysis, such as ECG data classification include LSTM, Bi-LSTM, CNN, with Bi-LSTM, and other sequential networks. However, these networks often struggle to accurately determine the long-range dependencies among data instances, which can result in problems such as vanishing or exploding gradients for longer data sequences. To address these shortcomings of sequential models, a hybrid arrhythmia classification system using recurrence along with a self-attention mechanism is developed. This system utilizes convolutional layers as a part of representation learning, designed to capture the salient features of raw ECG data. Then, the latent embedded layer is fed to a self-attention-assisted transformer encoder model. Because the ECG data are highly influenced by absolute order, position, and proximity of time steps due to interdependent relationships among immediate neighbors, a component of recurrence using Bi-LSTM is added to the encoder model to address this characteristic of the data. The model performance indices such as classification accuracy and F1-score were found to be 99.2%. This indicates that the combination of recurrence along with self-attention-assisted architecture produces improved classification of arrhythmia from raw ECG signal when compared with the state-of-the-art models. Full article

The topic of safety in chemistry laboratories in schools is crucial, as severe accidents in labs occur worldwide, primarily due to poorly trained individuals and improper behavior. One reason for this could be that the topic is often dry and boring for students. One solution to this problem is engaging students more actively in the lesson using a game format. In this publication, we present an augmented-reality-supported exit game in the form of a laboratory rally and the results of a pilot study that examined the use of the rally in terms of technology acceptance and intrinsic motivation. The study involved 22 students from a general high school. The study results show a high level of technology acceptance for the augmented reality used, as well as good results in terms of the intrinsic motivation triggered by the lesson. Full article

In this paper, an adaptive path construction approach for Mobile Sink (MS) in wireless sensor networks (WSNs) for data gathering has been proposed. The path is constructed based on selecting Rendezvous Points (RPs) in the sensing field where the MS stops in order to collect the data. Compared with the most existing RP-based schemes, which rely on fixed RPs to construct the path where these points will stay fixed during the whole network lifetime, we propose an adaptive path construction where the locations of the RPs are dynamically updated using a Fuzzy Inference System (FIS). The proposed FIS, which is named Fuzzy_RPs, has three inputs and one output. The inputs are: the remaining energy of the sensor nodes, the transmission distance between the RPs and the sensor nodes, and the number of surrounding neighbors of each node. The output of FIS is a weight value for each sensor node generated based on the previous three parameters and, thus, each RP is updated to its new location accordingly. Simulation results have shown that the proposed approach extends the network lifetime compared with another existing approach that uses fixed RPs. For example, in terms of using the first dead node as a metric for the network lifetime, when the number of deployed sensor nodes changes from 150 to 300, an improvement that ranges from 48.3% to 83.76% has been achieved compared with another related approach that uses fixed RPs. Full article

This paper describes a real case implementation of an automatic pedestrian-detection solution, implemented in the city of Aveiro, Portugal, using affordable LiDAR technology and open, publicly available, pedestrian-detection frameworks based on machine-learning algorithms. The presented solution makes it possible to anonymously identify pedestrians, and extract associated information such as position, walking velocity and direction in certain areas of interest such as pedestrian crossings or other points of interest in a smart-city context. All data computation (3D point-cloud processing) is performed at edge nodes, consisting of NVIDIA Jetson Nano and Xavier platforms, which ingest 3D point clouds from Velodyne VLP-16 LiDARs. High-performance real-time computation is possible at these edge nodes through CUDA-enabled GPU-accelerated computations. The MQTT protocol is used to interconnect publishers (edge nodes) with consumers (the smart-city platform). The results show that using currently affordable LiDAR sensors in a smart-city context, despite the advertising characteristics referring to having a range of up to 100 m, presents great challenges for the automatic detection of objects at these distances. The authors were able to efficiently detect pedestrians up to 15 m away, depending on the sensor height and tilt. Based on the implementation challenges, the authors present usage recommendations to get the most out of the used technologies. Full article

Industrial Internet of Things (IIoT) systems are enhancing the delivery of services and boosting productivity in a wide array of industries, from manufacturing to healthcare. However, IIoT devices are susceptible to cyber-threats such as the leaking of important information, products becoming compromised, and damage to industrial controls. Recently, blockchain technology has been used to increase the trust between stakeholders collaborating in the supply chain in order to preserve privacy, ensure the provenance of material, provide machine-led maintenance, etc. In all cases, such industrial blockchains establish a novel foundation of trust for business transactions which could potentially streamline and expedite economic processes to a significant extent. This paper presents an examination of “Schloss”, an industrial blockchain system architecture designed for multi-factory environments. It proposes an innovative solution to increase trust in industrial networks by incorporating a fairness concept as a subsystem of an industrial blockchain. The proposed mechanism leverages the concept of taxes imposed on blockchain nodes to enforce ethical conduct and discipline among participants. In this paper, we propose a game theory-based mechanism to address security and trust difficulties in industrial networks. The mechanism, inspired by the ultimatum game, progressively punishes malicious actors to increase the cost of fraud, improve the compensation system, and utilise the reward reporting capabilities of blockchain technology to further discourage fraudulent activities. Furthermore, the blockchain’s incentive structure is utilised to reduce collusion and speed up the process of reaching equilibrium, thereby promoting a secure and trustworthy environment for industrial collaboration. The objective of this paper is to address lack of trust among industrial partners and introduce a solution that brings security and trust to the forefront of industrial blockchain applications. Full article

The widespread use of technology has made communication technology an indispensable part of daily life. However, the present cloud infrastructure is insufficient to meet the industry’s growing demands, and multi-access edge computing (MEC) has emerged as a solution by providing real-time computation closer to the data source. Effective management of MEC is essential for providing high-quality services, and proactive self-healing is a promising approach that anticipates and executes remedial operations before faults occur. This paper aims to identify, evaluate, and synthesize studies related to proactive self-healing approaches in MEC environments. The authors conducted a systematic literature review (SLR) using four well-known digital libraries (IEEE Xplore, Web of Science, ProQuest, and Scopus) and one academic search engine (Google Scholar). The review retrieved 920 papers, and 116 primary studies were selected for in-depth analysis. The SLR results are categorized into edge resource management methods and self-healing methods and approaches in MEC. The paper highlights the challenges and open issues in MEC, such as offloading task decisions, resource allocation, and security issues, such as infrastructure and cyber attacks. Finally, the paper suggests future work based on the SLR findings. Full article

Next-generation networks are expected to handle a wide variety of internet of everything (IoE) services, notably including virtual reality (VR) for smart industrial-oriented applications. VR for industrial environments subtends strict quality of service constraints, requiring sixth-generation terahertz communications to be satisfied. In such an environment, an additional important issue is trying to get high utilization of network and computing resources. This implies identifying efficient access techniques and methodologies to increase bandwidth utilization and enable flows related to services, with different service requirements, to coexist on the same computation node. Towards this goal, this paper addresses the problem of coexistence of the traffic flows related to different services with given delay requirements, on the same computation node arranged to execute flow processing. In such a context, a theoretical comprehensive performance analysis, to the best of the authors’ knowledge, is still missing in the literature. As a consequence, this lack strongly limits the possibility of fully capturing the performance advantages of computation node sharing among different traffic flows, i.e., services. The proposed analysis aims to give a measure of the ability of the system in accomplishing services before the expiration of corresponding deadlines. The integration of martingale bounds within the stochastic network calculus tool is provided, assuming both the first-in–first-out and the earliest deadline first scheduling policies. Finally, the validity of the analysis proposed is confirmed by the tightness emerging from the comparison between analytical predictions and simulation results. Full article

Healthy and sufficient crop and food production are very much essential for everyone as the population is increasing globally. The production of crops affects the economy of a country to a great extent. In agriculture, observing the soil, weather, and water availability and, based on these factors, selecting an appropriate crop, finding the availability of seeds, analysing crop demand in the market, and having knowledge of crop cultivation are important. At present, many advancements have been made in recent times, starting from crop selection to crop cutting. Mainly, the roles of the Internet of Things, cloud computing, and machine learning tools help a farmer to analyse and make better decisions in each stage of cultivation. Once suitable crop seeds are chosen, the farmer shall proceed with seeding, monitoring crop growth, disease detection, finding the ripening stage of the crop, and then crop cutting. The main objective is to provide a continuous support system to a farmer so that he can obtain regular inputs about his field and crop. Additionally, he should be able to make proper decisions at each stage of farming. Artificial intelligence, machine learning, the cloud, sensors, and other automated devices shall be included in the decision support system so that it will provide the right information within a short time span. By using the support system, a farmer will be able to take decisive measures without fully depending on the local agriculture offices. We have proposed an IoT-enabled soil nutrient classification and crop recommendation (IoTSNA-CR) model to recommend crops. The model helps to minimise the use of fertilisers in soil so as to maximise productivity. The proposed model consists of phases, such as data collection using IoT sensors from cultivation lands, storing this real-time data into cloud memory services, accessing this cloud data using an Android application, and then pre-processing and periodic analysis of it using different learning techniques. A sensory system was prepared with optimised cost that contains different sensors, such as a soil temperature sensor, a soil moisture sensor, a water level indicator, a pH sensor, a GPS sensor, and a colour sensor, along with an Arduino UNO board. This sensory system allowed us to collect moisture, temperature, water level, soil NPK colour values, date, time, longitude, and latitude. The studies have revealed that the Agrinex NPK soil testing tablets should be applied to a soil sample, and then the soil colour can be sensed using an LDR colour sensor to predict the phosphorus (P), nitrogen (N), and potassium (K) values. These collected data together were stored in Firebase cloud storage media. Then, an Android application was developed to fetch and analyse the data from the Firebase cloud service from time to time by a farmer. In this study, a novel approach was identified via the hybridisation of algorithms. We have developed an algorithm using a multi-class support vector machine with a directed acyclic graph and optimised it using the fruit fly optimisation method (MSVM-DAG-FFO). The highest accuracy rate of this algorithm is 0.973, compared to 0.932 for SVM, 0.922 for SVM kernel, and 0.914 for decision tree. It has been observed that the overall performance of the proposed algorithm in terms of accuracy, recall, precision, and F-Score is high compared to other methods. The IoTSNA-CR device allows the farmer to maintain his field soil information easily in the cloud service using his own mobile with minimum knowledge. Additionally, it reduces the expenditure to balance the soil minerals and increases productivity. Full article

Currently, with the rapid development of deep learning, deep neural networks (DNNs) have been widely applied in various computer vision tasks. However, in the pursuit of performance, advanced DNN models have become more complex, which has led to a large memory footprint and high computation demands. As a result, the models are difficult to apply in real time. To address these issues, model compression has become a focus of research. Furthermore, model compression techniques play an important role in deploying models on edge devices. This study analyzed various model compression methods to assist researchers in reducing device storage space, speeding up model inference, reducing model complexity and training costs, and improving model deployment. Hence, this paper summarized the state-of-the-art techniques for model compression, including model pruning, parameter quantization, low-rank decomposition, knowledge distillation, and lightweight model design. In addition, this paper discusses research challenges and directions for future work. Full article

IPS is a crucial technology that enables medical staff and hospital management to accurately locate and track persons or assets inside medical buildings. Among other technologies, readily available BLE can be exploited to achieve an energy-efficient and low-cost solution. This work presents the design, implementation and comparison of a RSSI-based and a MCPD-based indoor localization system. The implementation is based on a lightweight wkNN algorithm that processes RSSI and MCPD distance data from connection-less BLE Beacons. The designed hardware and firmware are implemented around the state-of-the-art SoC for BLE, the nRF5340 from Nordic Semiconductor. Experimental evaluation with real-time data processing has been evaluated and presented in a 7.3 m × 8.9 m room with furniture and six beacon nodes. The experimental results on randomly chosen validation points within the room show an average error of only 0.50 m for the MCPD approach, whereas the RSSI approach achieved an error of 1.39 m. Full article

Parkinson’s disease (PD) is a serious movement disorder that may eventually progress to mild cognitive dysfunction (MCI) and dementia. According to the Parkinson’s foundation, one million Americans were diagnosed with PD and almost 10 million individuals suffer from the disease worldwide. An early and precise clinical diagnosis of PD will ensure an early initiation of effective therapeutic treatments, which will potentially slow down the progression of the disease and improve the quality of life for patients and their caregivers. Machine and deep learning are promising technologies that may assist and support clinicians in providing an objective and reliable diagnosis of the disease based upon significant and unique features identified from relevant medical data. In this paper, the author provides a comprehensive review of the artificial intelligence techniques that were recently proposed during the period from 2016 to 2022 for the screening and staging of PD as well as the identification of the biomarkers of the disease based on Electroencephalography (EEG), Magnetic Resonance Imaging (MRI), speech tests, handwriting exams, and sensory data. In addition, the author highlights the current and future trends for PD diagnosis based machine and deep learning and discusses the limitations, challenges, potential future solutions, and recommendations for a reliable application of machine and deep learning for PD detection and screening. Full article

Background: This study presents a graph-based, macro-scale, polarity-based, echo chamber detection approach for Twitter. Echo chambers are a concern as they can spread misinformation, and reinforce harmful stereotypes and biases in social networks. Methods: This study recorded the German-language Twitter stream over two months, recording about 6.7M accounts and their 75.5M interactions (33M retweets). This study focuses on retweet interaction patterns in the German-speaking Twitter stream and found that the greedy modularity maximization and HITS metric are the most effective methods for identifying echo chambers. Results: The purely structural detection approach identified an echo chamber (red community, 66K accounts) focused on a few topics with a triad of anti-Covid, right-wing populism and pro-Russian positions (very likely reinforced by Kremlin-orchestrated troll accounts). In contrast, a blue community (113K accounts) was much more heterogeneous and showed “normal” communication interaction patterns. Conclusions: The study highlights the effects of echo chambers as they can make political discourse dysfunctional and foster polarization in open societies. The presented results contribute to identifying problematic interaction patterns in social networks often involved in the spread of disinformation by problematic actors. It is important to note that not the content but only the interaction patterns would be used as a decision criterion, thus avoiding problematic content censorship. Full article

Technology-based in-home reading and spelling programs have the potential to compensate for the lack of sufficient instructions provided at schools. However, the recent COVID-19 pandemic showed the immaturity of the existing remote teaching solutions. Consequently, many students did not receive the necessary instructions. This paper presents a model for developing intelligent reading and spelling programs. The proposed approach is based on an optimization model that includes artificial neural networks and linear regression to maximize the educational value of the pedagogical content. This model is personalized, tailored to the learning ability level of each user. Regression models were developed for estimating the lexical difficulty in the literacy tasks of auditory and visual lexical decision, word naming, and spelling. For building these regression models, 55 variables were extracted from French lexical databases that were used with the data from lexical mega-studies. Forward stepwise analysis was conducted to identify the top 10 most important variables for each lexical task. The results showed that the accuracy of the models (based on root mean square error) reached 88.13% for auditory lexical decision, 89.79% for visual lexical decision, 80.53% for spelling, and 83.86% for word naming. The analysis of the results showed that word frequency was a key predictor for all the tasks. For spelling, the number of irregular phoneme-graphemes was an important predictor. The auditory word recognition depended heavily on the number of phonemes and homophones, while visual word recognition depended on the number of homographs and syllables. Finally, the word length and the consistency of initial grapheme-phonemes were important for predicting the word-naming reaction times. Full article

(1) Background: Augmented reality is no less popular than virtual reality. This technology has begun to be used in education fields, one of which is special education. Merging the real and virtual worlds is the advantage of augmented reality. However, it needs special attention in making software for children with special needs, such as children with autism. This paper presents an application prototype by paying attention to the characteristics of autistic individuals according to the Autism Guide, that has existed in previous studies. (2) Method: The method used in the development of this prototype is the Linear Sequential Model. Application development is made using Unity3D, Vuforia, and Adobe Illustrator by considering accessibility and other conveniences for developers. (3) Results: The prototype was developed with reference to the Autism Guide, then validated by media experts and autistic experts with the results of the assessment obtaining a score of 87.3/100 which is in the “Very Good” category and is suitable for use. (4) Conclusions: The development of a prototype that refers to the characteristics of children with autism needs to be considered so that what will be conveyed can be easily accepted. Full article

COVID-19 has raised the issue of fighting epidemics. We were able to realize that in this fight, countering the spread of the disease was the main goal and we propose to contribute to it. To achieve this, we propose an enriched model of Random Forest (RF) that we called RF EP (EP for Epidemiological Prediction). RF is based on the Forest RI algorithm, proposed by Leo Breiman. Our model (RF EP) is based on a modified version of Forest RI that we called Forest EP. Operations added on Forest RI to obtain Forest EP are as follows: the selection of significant variables, the standardization of data, the reduction in dimensions, and finally the selection of new variables that best synthesize information the algorithm needs. This study uses a data set designed for classification studies to predict whether a patient is suffering from COVID-19 based on the following 11 variables: Country, Age, Fever, Bodypain, Runny_nose, Difficult_in_breathing, Nasal_congestion, Sore_throat, Gender, Severity, and Contact_with_covid_patient. We compared default RF to five other machine learning models: GNB, LR, SVM, KNN, and DT. RF proved to be the best classifier of all with the following metrics: Accuracy (94.9%), Precision (94.0%), Recall (96.6%), and F1 Score (95.2%). Our model, RF EP, produced the following metrics: Accuracy (94.9%), Precision (93.1%), Recall (97.7%), and F1 Score (95.3%). The performance gain by RF EP on the Recall metric compared to default RF allowed us to propose a new model with a better score than default RF in the limitation of the virus propagation on the dataset used in this study. Full article

Academic discourse communities and learning circles are characterized by collaboration, sharing commonalities in terms of social interactions and language. The discourse of these communities is composed of jargon, common terminologies, and similarities in how they construe and communicate meaning. This study examines the extent to which discourse reveals “shared language” among its participants that can promote inclusion or affinity. Shared language is characterized in terms of linguistic features and lexical, syntactical, and semantic similarities. We leverage a multi-method approach, including (1) feature engineering using state-of-the-art natural language processing techniques to select the most appropriate features, (2) the bag-of-words classification model to predict linguistic similarity, (3) explainable AI using the local interpretable model-agnostic explanations to explain the model, and (4) a two-step cluster analysis to extract innate groupings between linguistic similarity and emotion. We found that linguistic similarity within and between the threaded discussions was significantly varied, revealing the dynamic and unconstrained nature of the discourse. Further, word choice moderately predicted linguistic similarity between posts within threaded discussions (accuracy = 0.73; F1-score = 0.67), revealing that discourse participants’ lexical choices effectively discriminate between posts in terms of similarity. Lastly, cluster analysis reveals profiles that are distinctly characterized in terms of linguistic similarity, trust, and affect. Our findings demonstrate the potential role of linguistic similarity in supporting social cohesion and affinity within online discourse communities. Full article

The GAME2AWE platform aims to provide a versatile tool for elderly fall prevention through exergames that integrate exercises, and simulate real-world environments and situations to train balance and reaction time using augmented and virtual reality technologies. In order to lay out the research area of interest, a review of the literature on systems that provide exergames for the elderly utilizing such technologies was conducted. The proposed use of augmented reality exergames on mobile devices as a complement to the traditional Kinect-based approach is a method that has been examined in the past with younger individuals in the context of physical activity interventions, but has not been studied adequately as an exergame tool for the elderly. An evaluation study was conducted with seniors, using multiple measuring scales to assess aspects such as usability, tolerability, applicability, and technology acceptance. In particular, the Unified Theory of Acceptance and Use of Technology (UTAUT) model was used to assess acceptance and identify factors that influence the seniors’ intentions to use the game platform in the long term, while the correlation between UTAUT factors was also investigated. The results indicate a positive assessment of the above user experience aspects leveraging on both qualitative and quantitative collected data. Full article

According to a study by the German Federal Printing Office (2022), every European lives with 90 digital identities on average, and the trend is rising. The German government has launched the innovation competition “Digital Identities Showcase” to select and promote identity projects for data security and sovereignty. The funding amount is 50 million EUR to develop software, research practical use cases and implement them by 2024. Of course, this large sum presupposes acceptance for the use of such digital identities, especially against the backdrop of critical opinions from the media and society, as already outlined in a Canadian study. However, there is little academic research on blockchain technology, but almost no article on the use of digital identities based on blockchain technology. This paper conducts a quantitative study on the social acceptance of digital identities using a questionnaire-based survey with 324 German participants on the social acceptance of the use of digital identities. The result of the study is that social acceptance of the use of digital identities is significantly influenced by demographics, citizens’ experience with blockchain products, affinity with financial products and privacy concerns. Full article

The aim of this research is to discuss the groundwork of building an Islamic Banking Document Screening Prototype based on a serverless architecture framework. This research first forms an algorithm for document matching based Vector Space Model (VCM) and adopts Levenshtein Distance for similarity setting. Product proposals will become a query, and policy documents by the central bank will be a corpus or database for document matching. Both the query and corpus went through preprocessing stage prior to similarity analysis. One set of queries with two sets of corpora is tested in this research to compare similarity values. Finally, a prototype of Shariah Document Screening is built based on a serverless architecture framework and ReactJS interface. This research is the first attempt to introduce a Shariah document screening prototype based on a serverless architecture technology that would be useful to the Islamic financial industry towards achieving a Shariah-compliant business. Given the development of Fintech, the output of this research study would be a complement to the existing Fintech applications, which focus on ensuring the Islamic nature of the businesses. Full article

The trend of E-commerce and online shopping is increasing rapidly. However, it is difficult to know about the quality of items from pictures and videos available on the online stores. Therefore, online stores and independent products reviews sites share user reviews about the products for the ease of buyers to find out the best quality products. The proposed work is about measuring and detecting product quality based on consumers’ attitude in product reviews. Predicting the quality of a product from customers’ reviews is a challenging and novel research area. Natural Language Processing and machine learning methods are popularly employed to identify product quality from customer reviews. Most of the existing research for the product review system has been done using traditional sentiment analysis and opinion mining. Going beyond the constraints of opinion and sentiment, such as a deeper description of the input text, is made possible by utilizing appraisal categories. The main focus of this study is exploiting the quality subcategory of the appraisal framework in order to predict the quality of the product. This paper presents a quality of product-based classification model (named QLeBERT) by combining quality of product-related lexicon, N-grams, Bidirectional Encoder Representations from Transformers (BERT), and Bidirectional Long Short Term Memory (BiLSTM). In the proposed model, the quality of the product-related lexicon, N-grams, and BERT are employed to generate vectors of words from part of the customers’ reviews. The main contribution of this work is the preparation of the quality of product-related lexicon dictionary based on an appraisal framework and automatically labelling the data accordingly before using them as the training data in the BiLSTM model. The proposed model is evaluated on an Amazon product reviews dataset. The proposed QLeBERT outperforms the existing state-of-the-art models by achieving an F1macro score of 0.91 in binary classification. Full article

Persona is a method to create a user profile by describing a fictitious user through user experience. This persona study needs to be carried out for the benefit of system design according to the users’ wishes because, so far, electronic resources (e-resources) are not widely used due to cognitive and affective factors such as limited subscription resources, limited user manuals, limited navigation features, and frequent errors when using electronic resources. This leaves the user feeling confused and stressed. The aim of this study is to obtain profiles of e-resource users in college libraries. The method used is an empathy map created with data from 32 users who answered questionnaires and participated in interviews. This study found four e-resource user personas in university libraries: lecturers, students, research assistants, and librarians. Users want a guide for using electronic resources that allows for chat and sharing, and which is fun and can be accessed from any device anytime and anywhere. The benefits of this study will be useful for designing e-resource systems according to users’ wishes. Full article

Currently, the primary concerns on the Internet are security and privacy, particularly in encrypted communications to prevent snooping and modification of Domain Name System (DNS) data by hackers who may attack using the HTTP protocol to gain illegal access to the information. DNS over HTTPS (DoH) is the new protocol that has made remarkable progress in encrypting Domain Name System traffic to prevent modifying DNS traffic and spying. To alleviate these challenges, this study explored the detection of DoH traffic tunnels of encrypted traffic, with the aim to determine the gained information through the use of HTTP. To implement the proposed work, state-of-the-art machine learning algorithms were used including Random Forest (RF), Gaussian Naive Bayes (GNB), Logistic Regression (LR), k-Nearest Neighbor (KNN), the Support Vector Classifier (SVC), Linear Discriminant Analysis (LDA), Decision Tree (DT), Adaboost, Gradient Boost (SGD), and LSTM neural networks. Moreover, ensemble models consisting of multiple base classifiers were utilized to carry out a series of experiments and conduct a comparative study. The CIRA-CIC-DoHBrw2020 dataset was used for experimentation. The experimental findings showed that the detection accuracy of the stacking model for binary classification was 99.99%. In the multiclass classification, the gradient boosting model scored maximum values of 90.71%, 90.71%, 90.87%, and 91.18% in Accuracy, Recall, Precision, and AUC. Moreover, the micro average ROC curve for the LSTM model scored 98%. Full article