Multi-class Approach Research Articles

An Efficient GAN-Based Multi-classification Approach for Financial Time Series Volatility Trend Prediction

Deep learning has achieved tremendous success in various applications owing to its robust feature representations of complex high-dimensional nonlinear data. Financial time-series prediction is no exception. Hence, the volatility trend prediction in financial time series (FTS) has been an active topic for several decades. Inspired by generative adversarial networks (GAN), which have been studied extensively in image processing and have achieved excellent results, we present the ordinal regression GAN for financial volatility trends (ORGAN-FVT) method for the end-to-end multi-classification task of FTS. An improved generative model based on convolutional long short-term memory (ConvLSTM) and multilayer perceptron (MLP) is proposed to capture temporal features effectively and mine the data distribution of volatility trends (short, neutral, and long) from given FTS data. Meanwhile, ordinal regression is leveraged for the discriminator to improve the multi-classification performance, making the model more practical. Finally, we empirically compare ORGAN-FVT with several state-of-the-art approaches on three real-world stock datasets: MICROSOFT(MSFT), Tesla(TSLA), and The People’s Insurance Company of China(PAICC). ORGAN-FVT demonstrated significantly better AUC and F1 scores, at most 20.81% higher than its competitors.

Feature Selection for Airbone LiDAR Point Cloud Classification

The classification of airborne LiDAR data is a prerequisite for many spatial data elaborations and analysis. In the domain of power supply networks, it is of utmost importance to be able to discern at least five classes for further processing—ground, buildings, vegetation, poles, and catenaries. This process is mainly performed manually by domain experts with the use of advanced point cloud manipulation software. The goal of this paper is to find a set of features which would divide space well enough to achieve accurate automatic classification on all relevant classes within the domain, thus reducing manual labor. To tackle this problem, we propose a single multi-class approach to classify all four basic classes (excluding ground) in a power supply domain with single pass-through, using one network. The proposed solution implements random forests and gradient boosting to create a feature-based per-point classifier which achieved an accuracy and F1 score of over 99% on all tested cases, with the maximum of 99.7% for accuracy and 99.5% for F1 score. Moreover, we achieved a maximum of 81.7% F1 score for the most sparse class. The results show that the proposed set of features for the LiDAR data cloud is effective in power supply line classification.

Deep Learning Multi-Class Approach for Human Fall Detection Based on Doppler Signatures.

Falling events are a global health concern with short- and long-term physical and psychological implications, especially for the elderly population. This work aims to monitor human activity in an indoor environment and recognize falling events without requiring users to carry a device or sensor on their bodies. A sensing platform based on the transmission of a continuous wave (CW) radio-frequency (RF) probe signal was developed using general-purpose equipment. The CW probe signal is similar to the pilot subcarriers transmitted by commercial off-the-shelf WiFi devices. As a result, our methodology can easily be integrated into a joint radio sensing and communication scheme. The sensing process is carried out by analyzing the changes in phase, amplitude, and frequency that the probe signal suffers when it is reflected or scattered by static and moving bodies. These features are commonly extracted from the channel state information (CSI) of WiFi signals. However, CSI relies on complex data acquisition and channel estimation processes. Doppler radars have also been used to monitor human activity. While effective, a radar-based fall detection system requires dedicated hardware. In this paper, we follow an alternative method to characterize falling events on the basis of the Doppler signatures imprinted on the CW probe signal by a falling person. A multi-class deep learning framework for classification was conceived to differentiate falling events from other activities that can be performed in indoor environments. Two neural network models were implemented. The first is based on a long-short-term memory network (LSTM) and the second on a convolutional neural network (CNN). A series of experiments comprising 11 subjects were conducted to collect empirical data and test the system's performance. Falls were detected with an accuracy of 92.1% for the LSTM case, while for the CNN, an accuracy rate of 92.1% was obtained. The results demonstrate the viability of human fall detection based on a radio sensing system such as the one described in this paper.

Deep Learning-Based Robust Automated System for Predicting Human Sperm DNA Fragmentation Index.

Determining the DNA fragmentation index (DFI) by the sperm chromatin dispersion (SCD) test involves manual counting of stained sperms with halo and no halo. The aim of this study is to build a robust artificial intelligence-based solution to predict the DFI. This is a retrospective experimental study conducted in a secondary in vitro fertilisation setup. We obtained 24,415 images from 30 patients after the SCD test using a phase-contrast microscope. We classified the dataset into two, binary (halo/no halo) and multiclass (big/medium/small halo/degraded (DEG)/dust). Our approach consists of a training and prediction phase. The 30 patients' images were divided into training (24) and prediction (6) sets. A pre-processing method M was developed to automatically segment the images to detect sperm-like regions and was annotated by three embryologists. To interpret the findings, the precision-recall curve and F1 score were utilised. Binary and multiclass datasets containing 8887 and 15,528 cropped sperm image regions showed an accuracy of 80.15% versus 75.25%. A precision-recall curve was determined and the binary and multiclass datasets obtained an F1 score of 0.81 versus 0.72. A confusion matrix was applied for predicted and actuals for the multiclass approach where small halo and medium halo confusion were found to be highest. Our proposed machine learning model can standardise and aid in arriving at accurate results without using expensive software. It provides accurate information about healthy and DEG sperms in a given sample, thereby attaining better clinical outcomes. The binary approach performed better with our model than the multiclass approach. However, the multiclass approach can highlight the distribution of fragmented and non-fragmented sperms.

Self-Adaptive Hybridized Lion Optimization Algorithm With Transfer Learning for Ancient Tamil Character Recognition in Stone Inscriptions

Tamil character recognition is an important field of research in pattern recognition and it is a technical challenge than other languages due to similarity and complexity of characters. Stone inscriptions reveal the details of lavishness, lifestyle, economic conditions, culture, and also of the managerial regulations followed by various rulers and dynasties particular to those regions. However, due to the long history of ancient stone inscription, natural erosion and lack of early protection measures, there are lot of noise in the existing ancient stone inscriptions, which create adverse effects on reading these stone inscriptions and their aesthetic appreciation. The research challenge in recognizing Tamil characters is mainly because of the characters with a number of holes, loops and curves. The number of letters in Tamil language is higher when compared to other languages. Even though there are various approaches provided by the researchers, challenges and issues still prevail in recognition of tamil text in stone inscriptions. In the existing systems, detection algorithms fail to produce desired accuracy and hence stone inscription recognition using transfer learning, a promising method is proposed in this paper. Lion Optimization Algorithm (LOA) is applied to optimize brightness and contrast and then stone inscription images are pre-processed for noise removal and then each character is separated by identifying contours. Characters are recognized using Transfer Learning (TL), a Deep Convolution Neural Network-based multi classification approach. The proposed hybrid model Self-Adaptive Lion Optimization Algorithm with Transfer Learning (SLOA-TL) when implemented in images of stone inscriptions achieves better accuracy and speed than other existing methods. It serves as an efficient design for recognition of tamil characters in stone inscriptions and preserving tamil traditional knowledge.

A Hybrid Method of Feature Extraction for Signatures Verification Using CNN and HOG a Multi-Classification Approach

The offline signature verification system’s feature extraction stage is regarded as crucial and has a big impact on how well these systems perform because the quantity and calibre of the features that are extracted determine how well these systems can distinguish between authentic and fake signatures. In this study, we introduced a hybrid method for extracting features from signature images, wherein Convolutional Neural Network (CNN) and Histogram of Oriented Gradients (HOG) were used, followed by the feature selection algorithm (Decision Trees) to identify the key features, and finally the CNN and HOG methods were combined. Three classifiers were employed to evaluate the hybrid method’s efficacy (long short-term memory, support vector machine, and K-nearest Neighbor). The experimental findings indicated that our suggested model executed quite satisfactorily in terms of efficiency and predictive ability, obtaining accuracy of (95.4%, 95.2%, and 92.7%) with UTSig dataset, and (93.7%, 94.1%, and 91.3%) with CEDAR dataset. This accuracy is deemed to be a high significance, particularly given that we checked skilled forged signatures that are more difficult to recognize than other forms of forged signatures like (simple or opposite-hand).

Automatic detection of brain tumors with the aid of ensemble deep learning architectures and class activation map indicators by employing magnetic resonance images

Today, as in every life-threatening disease, early diagnosis of brain tumors plays a life-saving role. The brain tumor is formed by the transformation of brain cells from their normal structures into abnormal cell structures. These formed abnormal cells begin to form in masses in the brain regions. Nowadays, many different techniques are employed to detect these tumor masses, and the most common of these techniques is Magnetic Resonance Imaging (MRI). In this study, it is aimed to automatically detect brain tumors with the help of ensemble deep learning architectures (ResNet50, VGG19, InceptionV3 and MobileNet) and Class Activation Maps (CAMs) indicators by employing MRI images. The proposed system was implemented in three stages. In the first stage, it was determined whether there was a tumor in the MR images (Binary Approach). In the second stage, different tumor types (Normal, Glioma Tumor, Meningioma Tumor, Pituitary Tumor) were detected from MR images (Multi-class Approach). In the last stage, CAMs of each tumor group were created as an alternative tool to facilitate the work of specialists in tumor detection. The results showed that the overall accuracy of the binary approach was calculated as 100% on the ResNet50, InceptionV3 and MobileNet architectures, and 99.71% on the VGG19 architecture. Moreover, the accuracy values of 96.45% with ResNet50, 93.40% with VGG19, 85.03% with InceptionV3 and 89.34% with MobileNet architectures were obtained in the multi-class approach.

ETECADx: Ensemble Self-Attention Transformer Encoder for Breast Cancer Diagnosis Using Full-Field Digital X-ray Breast Images.

Early detection of breast cancer is an essential procedure to reduce the mortality rate among women. In this paper, a new AI-based computer-aided diagnosis (CAD) framework called ETECADx is proposed by fusing the benefits of both ensemble transfer learning of the convolutional neural networks as well as the self-attention mechanism of vision transformer encoder (ViT). The accurate and precious high-level deep features are generated via the backbone ensemble network, while the transformer encoder is used to diagnose the breast cancer probabilities in two approaches: Approach A (i.e., binary classification) and Approach B (i.e., multi-classification). To build the proposed CAD system, the benchmark public multi-class INbreast dataset is used. Meanwhile, private real breast cancer images are collected and annotated by expert radiologists to validate the prediction performance of the proposed ETECADx framework. The promising evaluation results are achieved using the INbreast mammograms with overall accuracies of 98.58% and 97.87% for the binary and multi-class approaches, respectively. Compared with the individual backbone networks, the proposed ensemble learning model improves the breast cancer prediction performance by 6.6% for binary and 4.6% for multi-class approaches. The proposed hybrid ETECADx shows further prediction improvement when the ViT-based ensemble backbone network is used by 8.1% and 6.2% for binary and multi-class diagnosis, respectively. For validation purposes using the real breast images, the proposed CAD system provides encouraging prediction accuracies of 97.16% for binary and 89.40% for multi-class approaches. The ETECADx has a capability to predict the breast lesions for a single mammogram in an average of 0.048 s. Such promising performance could be useful and helpful to assist the practical CAD framework applications providing a second supporting opinion of distinguishing various breast cancer malignancies.

Multiclass spatial predictions of borehole yield in southern Mali by means of machine learning classifiers

Study regionRegions of Bamako, Kati and Kangaba, southwestern Mali Study focusMachine learning-based mapping of borehole yield. Three algorithms were trained on an imbalanced multiclass database of boreholes, while twenty variables were used as predictors for borehole yield. All models returned balanced and geometric scores in the order of 0.80, with area under the receiver operating characteristic curve up to 0.87. Three main methodological conclusions are drawn: (a) The evaluation of different machine learning classifiers and various resampling strategies and the subsequent selection of the best performing ones is shown to be a good strategy in this type of studies; (b) ad hoc calibration tools, such as data on borehole success rates, provide an apt complement to standard machine learning metrics; and (c) a multiclass approach with an unbalanced database represents a greater challenge than predicting a bivariate outcome, but potentially results in a finer depiction of field conditions. New hydrological insights for the regionAlluvial sediments were found to be the most productive areas, while the Mandingue Plateau has the lowest groundwater potential. The piedmont areas showcase an intermediate groundwater prospect. Elevation, basement depth, slope and geology rank among the most important variables. Lower values of clay content, slopes and elevations, and higher values of basement depth and saturated thickness were linked to the most productive class.

An intuitionistic fuzzy multi-class approach for risk assessment in sustainable BOT highway construction projects: A case study

An intuitionistic fuzzy multi-class approach for risk assessment in sustainable BOT highway construction projects: A case study

A Healthcare Paradigm for Deriving Knowledge Using Online Consumers' Feedback.

Home healthcare agencies (HHCAs) provide clinical care and rehabilitation services to patients in their own homes. The organization’s rules regulate several connected practitioners, doctors, and licensed skilled nurses. Frequently, it monitors a physician or licensed nurse for the facilities and keeps track of the health histories of all clients. HHCAs’ quality of care is evaluated using Medicare’s star ratings for in-home healthcare agencies. The advent of technology has extensively evolved our living style. Online businesses’ ratings and reviews are the best representatives of organizations’ trust, services, quality, and ethics. Using data mining techniques to analyze HHCAs’ data can help to develop an effective framework for evaluating the finest home healthcare facilities. As a result, we developed an automated predictive framework for obtaining knowledge from patients’ feedback using a combination of statistical and machine learning techniques. HHCAs’ data contain twelve performance characteristics that we are the first to analyze and depict. After adequate pattern recognition, we applied binary and multi-class approaches on similar data with variations in the target class. Four prominent machine learning models were considered: SVM, Decision Tree, Random Forest, and Deep Neural Networks. In the binary class, the Deep Neural Network model presented promising performance with an accuracy of 97.37%. However, in the case of multiple class, the random forest model showed a significant outcome with an accuracy of 91.87%. Additionally, variable significance is derived from investigating each attribute’s importance in predictive model building. The implications of this study can support various stakeholders, including public agencies, quality measurement, healthcare inspectors, and HHCAs, to boost their performance. Thus, the proposed framework is not only useful for putting valuable insights into action, but it can also help with decision-making.

A Multiclass Approach to Estimating Software Vulnerability Severity Rating with Statistical and Word Embedding Methods

The analysis and grading of software vulnerabilities is an important process that is done manually by experts today. For this reason, there are time delays, human errors, and excessive costs involved with the process. The final result of these software vulnerability reports created by experts is the calculation of a severity score and a severity rating. The severity rating is the first and foremost value of the software’s vulnerability. The vulnerabilities that can be exploited are only 20% of the total vulnerabilities. The vast majority of exploitations take place within the first two weeks. It is therefore imperative to determine the severity rating without time delays. Our proposed model uses statistical methods and deep learning-based word embedding methods from natural language processing techniques, and machine learning algorithms that perform multi-class classification. Bag of Words, Term Frequency Inverse Document Frequency and Ngram methods, which are statistical methods, were used for feature extraction. Word2Vec, Doc2Vec and Fasttext algorithms are included in the study for deep learning based Word embedding. In the classification stage, Naive Bayes, Decision Tree, K-Nearest Neighbors, Multi-Layer Perceptron, and Random Forest algorithms that can make multi-class classification were preferred. With this aspect, our model proposes a hybrid method. The database used is open to the public and is the most reliable data set in the field. The results obtained in our study are quite promising. By helping experts in this field, procedures will speed up. In addition, our study is one of the first studies containing the latest version of the data size and scoring systems it covers.

Android Spyware Detection Using Machine Learning: A Novel Dataset.

Smartphones are an essential part of all aspects of our lives. Socially, politically, and commercially, there is almost complete reliance on smartphones as a communication tool, a source of information, and for entertainment. Rapid developments in the world of information and cyber security have necessitated close attention to the privacy and protection of smartphone data. Spyware detection systems have recently been developed as a promising and encouraging solution for smartphone users’ privacy protection. The Android operating system is the most widely used worldwide, making it a significant target for many parties interested in targeting smartphone users’ privacy. This paper introduces a novel dataset collected in a realistic environment, obtained through a novel data collection methodology based on a unified activity list. The data are divided into three main classes: the first class represents normal smartphone traffic; the second class represents traffic data for the spyware installation process; finally, the third class represents spyware operation traffic data. The random forest classification algorithm was adopted to validate this dataset and the proposed model. Two methodologies were adopted for data classification: binary-class and multi-class classification. Good results were achieved in terms of accuracy. The overall average accuracy was 79% for the binary-class classification, and 77% for the multi-class classification. In the multi-class approach, the detection accuracy for spyware systems (UMobix, TheWiSPY, MobileSPY, FlexiSPY, and mSPY) was 90%, 83.7%, 69.3%, 69.2%, and 73.4%, respectively; in binary-class classification, the detection accuracy for spyware systems (UMobix, TheWiSPY, MobileSPY, FlexiSPY, and mSPY) was 93.9%, 85.63%, 71%, 72.3%, and 75.96%; respectively.

Multi-temporal Satellite Images and Combination of Multi-Classification Approaches for Evaluation of Changes in Forest Cover: A case study

The primary objective of this study is to develop a method to monitor the changes in forest cover using three different techniques, namely Spatial Data Mining (SDM), supervised and unsupervised classification, and Geographical Information System (GIS) overlay processing methods. The secondary objective was to monitor the land use changes using multi-temporal satellite images (Landsat 8 images of 2016 and 2020). The study area is located in Nuwara Eliya, Sri Lanka. The first and second approaches consisted of supervised and unsupervised classification approaches. The use of a combination of two classification mapping approaches provides better results compared with the use of a single classification, which is a novelty of the present study. The GIS overly processing technique combined the two maps obtained from supervised and unsupervised classifications to develop an improved map for land use changes. The improved map was reclassified and converted into ASCII (American Standard Code for Information Interchange) format, which was then pre-processed to convert the data set into the suitable format for the SDM modeling. The converted format was used to implement SDM modeling with the support of the Clustering-outline detection algorithm. The overall accuracy of the Landsat 8 images was 94.6. Results revealed that the forest cover extent was diminished by 5.28% in the study area between 2016 and 2020. The ground measurement was done with the help of the forest department to verify the results. The study revealed that the forest area decreased, and farming lands increased due to intensive agriculture. Future studies would be necessary to determine the validity and suitability of the developed model for other climatic zones in the country.

Ensemble Filter technique for Detection and Classification of attacks in Cloud Computing

In all technologies, including traditional computing and cloud computing, security has always been the primary concern. In recent years, cloud computing has become widely accepted on a global scale. Cyber attacks aimed at it have increased along with its widespread acceptance. Although ample research is done in the security domain and cloud computing is based on rigid security fundamentals, the advancing network security attacks create the need for an advanced security mechanism. Also, the multiclass classification strategy has received very little attention, and classification accuracy can yet be improved. Hence, this work proposes an Ensemble Filter-based Intrusion Detection System (EFIDS) to address the limitations of previous research work. It not only identifies malicious traffic but also categorizes the attempted attacks (multiclass classification). The famous intrusion detection benchmark dataset, NSL KDD, is used to evaluate the model. Using the model, it was possible to enhance the classification accuracy of both binary and multiclass approaches up to 99.85 percent and 99.63 percent, respectively. Additionally, both forms of classification have shown a 65–70% improvement in training time.

A New Hybrid Multiclass Approach Based on KNN and SVM

Support vector machine (SVM) is a machine learning method widely used in solving binary data classification problems due to its performance. Nevertheless, in practical problems of classification, there are often cases of the presence of more than two classes of objects in the original dataset. The paper considers a solution to the problem of SVM multiclass with the aim to increase the data classification quality based on a new way of hybridisation between SVM and [Formula: see text]-nearest neighbour (KNN) algorithms. The first phase of the approach is called the filtering phase. At this level, the feature space is split into two classes separated by a hyperplane. In the next step called review, we generate a second hyperplane, then we calculate the distance between each test pattern and the second hyperplane in the feature space using e.g. the KNN function. The result of the two phases is three classes instead of two produced by the conventional SVM. For evaluation purposes, dataset experiments are conducted on seven benchmark datasets that have high dimensionality and large size. Numerical experiments show that the 3SVM approach can improve not only the accuracy compared to other multiclass SVM approaches, but also the precision, recall, and [Formula: see text]-score.

Multi-Label Attribute Selection of Arrhythmia for Electrocardiogram Signals with Fusion Learning.

There are three primary challenges in the automatic diagnosis of arrhythmias by electrocardiogram (ECG): the significant variation among individual patients, the multiple pathologies in the ECG signal and the high cost in annotating clinical ECG with the corresponding labels. Traditional ECG processing approaches rely heavily on prior knowledge, such as those from feature extraction and waveform analysis. The preprocessing for prior knowledge incurs computational overhead. Furthermore, standard deep learning methods do not fully consider the dynamic temporal, spatial and multi-labeling characteristics of ECG data. In clinical ECG waveforms, it is common to see multi-labeling in which a patient is labeled with multiple classes of arrhythmias. However, multiclass approaches in current research mainly solve the multi-label machine learning problem, ignoring the correlation between diseases, resulting in information loss. In this paper, an arrhythmia detection and classification scheme called multi-label fusion deep learning is proposed. The objective is to build a unified system with automatic feature learning which supports effective multi-label classification. First, a multi-label ECG-based feature selection method is combined with a matrix decomposition and sparse learning theory. The optimal feature subset is selected as a preprocessing algorithm for ECG data. A multi-label classifier is then constructed by fusing CNN and RNN networks to fully exploit the interactions and features of the time and space dimensions. The experimental result demonstrates that the proposed method can achieve a state-of-the-art performance compared to other algorithms in multi-label database experiments.

Social media data analysis framework for disaster response

This paper presents a social media data analysis framework applied to multiple datasets. The method developed uses machine learning classifiers, where filtering binary classifiers based on deep bidirectional neural networks are trained on benchmark datasets of disaster responses for earthquakes and floods and extreme flood events. The classifiers consist of learning from discrete handcrafted features and fine-tuning approaches using deep bidirectional Transformer neural networks on these disaster response datasets. With the development of the multiclass classification approach, we compare the state-of-the-art results in one of the benchmark datasets containing the largest number of disaster-related categories. The multiclass classification approaches developed in this research with support vector machines provide a precision of 0.83 and 0.79 compared to Bernoulli naïve Bayes, which are 0.59 and 0.76, and multinomial naïve Bayes, which are 0.79 and 0.91, respectively. The binary classification methods based on the MDRM dataset show a higher precision with deep learning methods (DistilBERT) than BoW and TF-IDF, while in the case of UnifiedCEHMET dataset show a high performance for accuracy with the deep learning method in terms of severity, with a precision of 0.92 compared to BoW and TF-IDF method which has a precision of 0.68 and 0.70, respectively.

A Novel Detection and Multi-Classification Approach for IoT-Malware Using Random Forest Voting of Fine-Tuning Convolutional Neural Networks.

The Internet of Things (IoT) is prone to malware assaults due to its simple installation and autonomous operating qualities. IoT devices have become the most tempting targets of malware due to well-known vulnerabilities such as weak, guessable, or hard-coded passwords, a lack of secure update procedures, and unsecured network connections. Traditional static IoT malware detection and analysis methods have been shown to be unsatisfactory solutions to understanding IoT malware behavior for mitigation and prevention. Deep learning models have made huge strides in the realm of cybersecurity in recent years, thanks to their tremendous data mining, learning, and expression capabilities, thus easing the burden on malware analysts. In this context, a novel detection and multi-classification vision-based approach for IoT-malware is proposed. This approach makes use of the benefits of deep transfer learning methodology and incorporates the fine-tuning method and various ensembling strategies to increase detection and classification performance without having to develop the training models from scratch. It adopts the fusion of 3 CNNs, ResNet18, MobileNetV2, and DenseNet161, by using the random forest voting strategy. Experiments are carried out using a publicly available dataset, MaleVis, to assess and validate the suggested approach. MaleVis contains 14,226 RGB converted images representing 25 malware classes and one benign class. The obtained findings show that our suggested approach outperforms the existing state-of-the-art solutions in terms of detection and classification performance; it achieves a precision of 98.74%, recall of 98.67%, a specificity of 98.79%, F1-score of 98.70%, MCC of 98.65%, an accuracy of 98.68%, and an average processing time per malware classification of 672 ms.

A Multi-Class Approach for the Automatic Detection of Congestive Heart Failure in Windowed ECG.

Congestive heart failure (CHF) is a chronic heart disease that causes debilitating symptoms and leads to higher mortality and morbidity. In this paper, we present HARPER, a novel automatic detector of CHF episodes able to distinguish between Normal Sinus Rhythm (NSR), CHF, and no-CHF. The main advantages of HARPER are its reliability and its capability of providing an early diagnosis. Indeed, the method is based on evaluating real-time features and observing a brief segment of ECG signal. HARPER is an independent tool meaning that it does not need any ECG annotation or segmentation algorithms to provide detection. The approach was submitted to complete experimentation by involving both the intra- and inter-patient validation schemes. The results are comparable to the state-of-art methods, highlighting the suitability of HARPER to be used in modern IoMT systems as a multi-class, fast, and highly accurate detector of CHF. We also provide guidelines for configuring a temporal window to be used in the automatic detection of CHF episodes.