Comparison of machine learning algorithms with regression analysis to predict the COVID-19 outbreak in Thailand

Coronavirus disease (COVID-19) is a public health problem in Thailand. Currently, there are more than 5 million infected people and the rate has been increasing at some point. It is therefore important to forecast the number of new cases over a short period of time to assist in strategic planning for the response to COVID-19. The purpose of this research paper was to compare the efficiency and prediction of the number of COVID-19 cases in Thailand using machine learning of 8 models using a regression analysis method. Using the 475-day dataset of COVID-19 cases in Thailand, the results showed that the predictive accuracy model (R2 score) from the testing dataset was the random forest (RF) model, which was 99.06%, followed by K-nearest neighbor (KNN), XGBoost. And the decision tree (DT) had the precision of 98.97, 98.67, and 98.64, respectively. And the results of the comparison of the number of infected people obtained from the prediction The models that predicted the number of real infections were the decision tree, random forest, and XGBoost, which were effective at predicting the number of infections correctly in the 2-4 day period. © 2023 Institute of Advanced Engineering and Science. All rights reserved.

Prognostic significance of chest CT severity score in mortality prediction of COVID-19 patients, a machine learning study

Background: The high mortality rate of COVID-19 makes it necessary to seek early identification of high-risk patients with poor prognoses. Although the association between CT-SS and mortality of COVID-19 patients was reported, its prognosis significance in combination with other prognostic parameters was not evaluated yet. Method(s): This retrospective single-center study reviewed a total of 6854 suspected patients referred to Imam Khomeini hospital, Ilam city, west of Iran, from February 9, 2020 to December 20, 2020. The prognostic performances of k-Nearest Neighbors (kNN), Multilayer Perceptron (MLP), Support Vector Machine (SVM), and J48 decision tree algorithms were evaluated based on the most important and relevant predictors. The metrics derived from the confusion matrix were used to determine the performance of the ML models. Result(s): After applying exclusion criteria, 815 hospitalized cases were entered into the study. Of these, 447(54.85%) were male and the mean (+/- SD) age of participants was 57.22(+/- 16.76) years. The results showed that the performances of the ML algorithms were improved when they are fed by the dataset with CT-SS data. The kNN model with an accuracy of 94.1%, sensitivity of 100. 0%, precision of 89.5%, specificity of 88.3%, and AUC around 97.2% had the best performance among the other three ML techniques. Conclusion(s): The integration of CT-SS data with demographics, risk factors, clinical manifestations, and laboratory parameters improved the prognostic performances of the ML algorithms. An ML model with a comprehensive collection of predictors could identify high-risk patients more efficiently and lead to the optimal use of hospital resources.Copyright © 2023, The Author(s).

Machine Learning Price Prediction During and Before COVID-19 and Consumer Buying Behavior

In the unprecedented situation of COVID-19, the global economy has turned upside down. This has led to sudden and unprecedented pressures on products demand and price forecasting. The study utilized regression techniques to predict product prices during and before COVID-19 using multiple influencing factors such as increase of COVID-19 positive cases on daily basis, number of deaths on a particular day, and government restrictions level. The data was gathered from worldometers website and combined with local store on sales based on the date. The results were eye opening as the product sold in the months of Mach, April, May and June 2020 were different than last year. This means the customers buying habits were totally altered due to many reasons such as job loss, wages reduction due to remote working, or promotions. Moreover, these products prices were directly proportional to increase of new COVID-19 cases, rise of daily deaths and government restriction levels imposed during the pandemic. The study uses machine learning data mining algorithms such as Logistic regression (LR), Decision Tree, Random Forest and K-Nearest Neighbor. Decision Tree and Random Forest works best in the pandemic situation to predict product price as compared to Logistic Regression and KNN. However, different outcomes were recorded when comparing the sales during pandemic and before pandemic. © 2023, The Author(s), under exclusive license to Springer Nature Switzerland AG.

IoT-cloud based healthcare model for COVID-19 detection: an enhanced k-Nearest Neighbour classifier based approach

COVID - 19 affected severely worldwide. The pandemic has caused many causalities in a very short span. The IoT-cloud-based healthcare model requirement is utmost in this situation to provide a better decision in the covid-19 pandemic. In this paper, an attempt has been made to perform predictive analytics regarding the disease using a machine learning classifier. This research proposed an enhanced KNN (k NearestNeighbor) algorithm eKNN, which did not randomly choose the value of k. However, it used a mathematical function of the dataset's sample size while determining the k value. The enhanced KNN algorithm eKNN has experimented on 7 benchmark COVID-19 datasets of different size, which has been gathered from standard data cloud of different countries (Brazil, Mexico, etc.). It appeared that the enhanced KNN classifier performs significantly better than ordinary KNN. The second research question augmented the enhanced KNN algorithm with feature selection using ACO (Ant Colony Optimization). Results indicated that the enhanced KNN classifier along with the feature selection mechanism performed way better than enhanced KNN without feature selection. This paper involves proposing an improved KNN attempting to find an optimal value of k and studying IoT-cloud-based COVID - 19 detection. [ABSTRACT FROM AUTHOR] Copyright of Computing is the property of Springer Nature and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Gabor Filter-Based Caries Image Feature Analysis Using Machine Learning

Dental caries sufferers in Indonesia demonstrate a higher frequency than other dental diseases even before the Covid-19 pandemic. The high risk of spreading the virus during the pandemic hinders handling dental care patients. Teledentistry is suggested as the main alternative to reduce the risk of spreading the virus. This study aims to establish a system for classifying the level of dental caries based on texture applicable for clinical implementation. Dental caries images were extracted using the Gabor Filter method and classified using the Support Vector Machine (SVM) and K-Nearest Neighbor (K-NN). A downsampling technique was applied to this system to reduce the large number of features affecting the classification time. System testing revealed that the Cubic SVM model generated the best result: Accuracy of 90.5%, precision of 89.75%, recall of 89.25%, specificity of 91.75%, and f-score of 88.5%. © 2022 IEEE.

Machine Learning Models Towards Prediction of COVID and Non-COVID 19 Patients in the Hospital's Intensive Care Units (ICU)

The global proliferation of COVID-19 prompted research towards the virus's detection and eventual eradication. One important area of research is the use of machine learning (ML) to realize and battle COVID-19. The goal of this study is to use machine learning to monitor COVID and non-COVID-19 patients and decide whether or not to transfer them to the intensive care unit (ICU). The precise disease diagnosis was essential due to the lack of oxygen supplementation in the majority of hospitals around the world. It will improve the effectiveness of the ICU facilities and lessen the load on the medical personnel and the ICU facilities by accurately forecasting how patients will be treated. If stable patients are recognized among all patients, home treatment could be established for stable patients. In this research, three machine learning algorithms were chosen as the method used, which are K-Nearest Neighbor (KNN), Support Vector Machine (SVM), and Extra Tree Classifier. These algorithms were chosen for their simplicity and robustness and based on the conducted literature review. A dataset containing 100 ICU and 131 stable patients of Covid and non-Covid samples from 24th Moscow City State Hospital was used. By using SMOTE technique with 10-fold cross-validation and feature selection on the dataset, KNN achieved an accuracy of 94.65%, SVM with an accuracy of 94.65%, and an accuracy of 96.18% for the Extra Tree Classifier. The outcomes of this research on the selected dataset prove how accurate these algorithms were able to predict the classes © 2022, Mathematical Modelling of Engineering Problems.All Rights Reserved.

Comparison of K-Nearest Neighbor and Logistic Regression Algorithms on Sentiment Analysis of Covid-19 Vaccination on Twitter with Vader And Textblob Labeling

In this analysis, the methods used are the K-Nearest Neighbor classification method and the Logistic Regression classification method with data taken on the twitter application. This study examines the level of accuracy in public sentiment regarding covid-19 vaccination with positive and negative labels. The AUC value in the KNN algorithm with TextBlob labeling is 0.765 with and 0.76S for VaderSentiment labeling are both included in the fair classification criteria. Meanwhile, the Logistic Regression algorithm produces an accuracy of 84.97% with a ratio of 90:10 for Labeling TextBlob, while for Labeling VaderSentiment with a ratio of 90:10 results in an accuracy of 85.22%. Both algorithms are validated using K-Fold Cross Validation with a fold count of 10. The comparison results obtained when conducting an evaluation with the confusion matrix showed that the Logistic Regression algorithm with VaderSentiment labeling had the highest accuracy value compared to the K-Nearest Neighbor algorithm with TextBlob and VaderSentiment labeling. © 2022 IEEE.

Application of Machine Learning in the Fight against the COVID-19 Pandemic: A Review

Introduction: Machine learning (ML) plays a significant role in the fight against the COVID-19 (officially known as SARS-CoV-2) pandemic. ML techniques enable the rapid detection of patterns and trends in large datasets. Therefore, ML provides efficient methods to generate knowledge from structured and unstructured data. This potential is particularly significant when the pandemic affects all aspects of human life. It is necessary to collect a large amount of data to identify methods to prevent the spread of infection, early detection, reduction of consequences, and finding appropriate medicine. Modern information and communication technologies (ICT) such as the Internet of Things (IoT) allow the collection of large amounts of data from various sources. Thus, we can create predictive ML-based models for assessments, predictions, and decisions. Method(s): This is a review article based on previous studies and scientifically proven knowledge. In this paper, bibliometric data from authoritative databases of research publications (Web of Science, Scopus, PubMed) are combined for bibliometric analyses in the context of ML applications for COVID-19. Aim(s): This paper reviews some ML-based applications used for mitigating COVID-19. We aimed to identify and review ML potentials and solutions for mitigating the COVID-19 pandemic as well as to present some of the most commonly used ML techniques, algorithms, and datasets applied in the context of COVID-19. Also, we provided some insights into specific emerging ideas and open issues to facilitate future research. Conclusion(s): ML is an effective tool for diagnosing and early detection of symptoms, predicting the spread of a pandemic, developing medicines and vaccines, etc.Copyright © 2022 Sciendo. All rights reserved.

Machine learning-assisted remote patient monitoring with data analytics

The health condition of the patients needs to be monitored with immense care. Healthcare promotes good health, helps in monitoring the patient's health status, disease diagnosis, and its management along with recovery. Monitoring the health condition postdischarge or postoperation is required to ensure a speedy recovery. Healthcare services can benefit from technological advancements to ensure better service. Healthcare assisted with machine learning techniques plays a significant role in the effective diagnosis of ailments, monitoring patient's health condition, and extend support in taking suitable measures during abnormality. In the proposed work, we collect the patient's data using sensors and upload them to the cloud. The collected data are subjected to preprocessing followed by analysis. The patient's health is remotely monitored, and machine learning techniques are applied to foretell abnormalities in the patient's health condition. Existing remote monitoring systems are not flexible and, hence, may result in an increased number of false positives. We try to reduce unnecessary alerts via machine learning methods and data analytics. Essential attributes like pulse rate, blood pressure, temperature, gender, and cholesterol levels of the patient are taken into consideration while predicting the results. In the time of pandemics, like COVID-19 with the scarce availability of medical personnel and treatment resources, this prediction may help in taking appropriate measures at the earliest. We train the model with the Kaggle Heart Disease UCI data set and test the model with real-time patient data. We apply our model to k nearest neighbor (KNN) and Naïve Bayes algorithm. The KNN has performed well over the Naïve Bayes algorithm. © 2022 Scrivener Publishing LLC.

Dynamic Candidate Solution Boosted Beluga Whale Optimization Algorithm for Biomedical Classification

In many fields, complicated issues can now be solved with the help of Artificial Intelligence (AI) and Machine Learning (ML). One of the more modern Metaheuristic (MH) algorithms used to tackle numerous issues in various fields is the Beluga Whale Optimization (BWO) method. However, BWO has a lack of diversity, which could lead to being trapped in local optimaand premature convergence. This study presents two stages for enhancing the fundamental BWO algorithm. The initial stage of BWO's Opposition-Based Learning (OBL), also known as OBWO, helps to expedite the search process and enhance the learning methodology to choose a better generation of candidate solutions for the fundamental BWO. The second step, referred to as OBWOD, combines the Dynamic Candidate Solution (DCS) and OBWO based on the k-Nearest Neighbor (kNN) classifier to boost variety and improve the consistency of the selected solution by giving potential candidates a chance to solve the given problem with a high fitness value. A comparison study with present optimization algorithms for single-objective bound-constraint optimization problems was conducted to evaluate the performance of the OBWOD algorithm on issues from the 2022 IEEE Congress on Evolutionary Computation (CEC'22) benchmark test suite with a range of dimension sizes. The results of the statistical significance test confirmed that the proposed algorithm is competitive with the optimization algorithms. In addition, the OBWOD algorithm surpassed the performance of seven other algorithms with an overall classification accuracy of 85.17% for classifying 10 medical datasets with different dimension sizes according to the performance evaluation matrix.

Application of machine learning approaches to predict the impact of ambient air pollution on outpatient visits for acute respiratory infections

With a remarkable increase in industrialization among fast-developing countries, air pollution is rising at an alarming rate and has become a public health concern. The study aims to examine the effect of air pollution on patient's hospital visits for respiratory diseases, particularly Acute Respiratory Infections (ARI). Outpatient hospital visits, air pollution and meteorological parameters were collected from March 2018 to October 2021. Eight machine learning algorithms (Random Forest model, K-Nearest Neighbors regression model, Linear regression model, LASSO regression model, Decision Tree Regressor, Support Vector Regression, X.G. Boost and Deep Neural Network with 5-layers) were applied for the analysis of daily air pollutants and outpatient visits for ARI. The evaluation was done by using 5-cross-fold confirmations. The data was randomly divided into test and training data sets at a scale of 1:2, respectively. Results show that among the studied eight machine learning models, the Random Forest model has given the best performance with R2 = 0.606, 0.608 without lag and 1-day lag respectively on ARI patients and R2 = 0.872, 0.871 without lag and 1-day lag respectively on total patients. All eight models did not perform well with the lag effect on the ARI patient dataset but performed better on the total patient dataset. Thus, the study did not find any significant association between ARI patients and ambient air pollution due to the intermittent availability of data during the COVID-19 period. This study gives insight into developing machine learning programs for risk prediction that can be used to predict analytics for several other diseases apart from ARI, such as heart disease and other respiratory diseases. © 2022 Elsevier B.V.

Diagnosis of Skin Cancer by Using Fuzzy-Ann Expert System with Unification of Improved Gini Index Random Forest-Based Feature

In addition to being one of the most widespread and lethal diseases in the world, skin cancer is also one of the most common types of cancer. However, due to its complexity and fuzzy nature, the clinical diagnosis process of any disease, including skin cancer, prostate cancer, coronary artery disorders, diabetes, and COVID-19, is frequently accompanied by doubt. In order to address the uncertainty and ambiguity surrounding the diagnosis of skin cancer as well as the heavier burden on the overlay of the network nodes of the fuzzy neural network system that frequently occurs due to insignificant features that are used to predict or diagnose the disease, a fuzzy neural network expert system with an improved Gini index random forest-based feature importance measure algorithm was proposed in this work. A Greater Gini Index Out of the 30 features in the dataset, the five most fitting features of the diagnostic Wisconsin breast cancer database were chosen using a random forest-based feature importance measure algorithm. Two sets of classification models were created using the logistic regression, support vector machine, k-nearest neighbour, random forest, and Gaussian naive Bayes learning algorithms. As a result, models for classification that included all features (30) and models that only used the top five features were used. The efficacy of the two sets of categorization models was assessed, and the results of the assessment were compared. The comparison's results show that the models with the fittest features outperformed those with the most complete features in terms of accuracy, sensitivity, and sensitivity. A fuzzy neural network-based expert system was therefore developed, utilising the five best features, and it achieved 99.83 percent accuracy, 99.86 percent sensitivity, and 99.64 percent specificity. The system built in this study also stands to be the best in terms of accuracy, sensitivity, and specificity when compared to prior research that used fuzzy neural networks or other applicable artificial intelligence techniques on the same dataset for the diagnosis of skin cancer. The z-test was also performed, and the test result demonstrates that the system has significantly improved accuracy for early skin cancer diagnosis. Copyright © 2023 Wolters Kluwer Medknow Publications. All rights reserved.

A hybrid CNN-KNN approach for identification of COVID-19 with 5-fold cross validation.

The novel coronavirus is the new member of the SARS family, which can cause mild to severe infection in the lungs and other vital organs like the heart, kidney and liver. For detecting COVID-19 from images, traditional ANN can be employed. This method begins by extracting the features and then feeding the features into a suitable classifier. The classification rate is not so high as feature extraction is dependent on the experimenters' expertise. To solve this drawback, a hybrid CNN-KNN-based model with 5-fold cross-validation is proposed to classify covid-19 or non-covid19 from CT scans of patients. At first, some pre-processing steps like contrast enhancement, median filtering, data augmentation, and image resizing are performed. Secondly, the entire dataset is divided into five equal sections or folds for training and testing. By doing 5-fold cross-validation, the generalization of the dataset is ensured and the overfitting of the network is prevented. The proposed CNN model consists of four convolutional layers, four max-pooling layers, and two fully connected layers combined with 23 layers. The CNN architecture is used as a feature extractor in this case. The features are taken from the CNN model's fourth convolutional layer and finally, the features are classified using K Nearest Neighbor rather than softmax for better accuracy. The proposed method is conducted over an augmented dataset of 4085 CT scan images. The average accuracy, precision, recall and F1 score of the proposed method after performing a 5-fold cross-validation is 98.26%, 99.42%,97.2% and 98.19%, respectively. The proposed method's accuracy is comparable with the existing works described further, where the state of the art and the custom CNN models were used. Hence, this proposed method can diagnose the COVID-19 patients with higher efficiency.

GAUGING STRESS, ANXIETY, DEPRESSION IN STUDENT DURING COVID-19 PANDEMIC

During the beginning of COVID-19 pandemic, studies came across the world concerning with health issues. Researches began to find the repercussions of the virus. The virus was found to be versatile as it changes its nature and targets the lungs of a person. Later, it was seen an astonishing massacre around the world due to the virus. Many people have lost their life but many more people are still suffering with bad psychological state. Researchers began to research on the nature virus but very few researches were made on the other side-effects of this pandemic. One such crucial subject to attend in contemporary world is the effect of COVID-19 on psychological state in general population. This side-effect may lead to raise an alarming situation in future that could result in more death cases. The proposed paper presents a study on the detection of stress and depression in people caused by the pandemic. The proposed methodology is based on perceived questionnaire method through which people's responses are recorded in the form of text. COVID victims have been interrogated against a set of questions and their responses are recorded. The methodology performs text mining of their responses that also include the people's reaction from social networking sites. The text processing of people's responses is done by natural language processing (NLP). NLP is used to interpret textural facts into meaningful segments that must be understandable to machine. The refined data has been transformed into PSS (perceived stress scale) scaling factor that ranges from 0 to 4 showing various level of stress. The proposed system utilized artificial intelligence in which na'́ive Bayes classifier, K-nearest neighbor (KNN), Decision tree and Random forest algorithms are applied to predict the emotional state of a person. The proposed system also uses data from social networking site for testing purpose. The model successfully shows a comparative study of such three classifiers for the classification of stress level into stress, anxiety and depression © 2022 SCPE

Passenger Flow Prediction Method for Rail Transit Stations Based on Empirical Mode Decomposition and K-nearest Neighbors

A passenger flow time series forecasting method based on empirical mode decomposition (EMD) and K-nearest neighbor nonparametric regression (KNN) was proposed. Based on the principle of EMD and KNN algorithm, the EMD-KNN combined algorithm flow was constructed on the basis of improving KNN prediction method. According to the characteristics that the time series trend of passenger flow has changed obviously due to the influence of COVID-19 epidemic situation in the example stations. BP structural breakpoint detection method was used to identify three structural breakpoints, and the time series segment with the closest passenger flow change trend to the forecast day was selected for empirical mode decomposition. The decomposed sequences were reorganized into high-frequency, low-frequency and trend sequences, and then the K-nearest neighbor algorithm considering weight was used to predict, and the final prediction results were obtained by superposition, and compared with the prediction results of single KNN algorithm and ARIMA model. The results show that the prediction accuracy of EMD-KNN combination algorithm is higher than that of single KNN algorithm and ARIMA model, and it can effectively capture the changing trend of passenger flow. © 2022, Editorial Department of Journal of Wuhan University of Technology. All right reserved.

COVID19-MLSF: A multi-task learning-based stock market forecasting framework during the COVID-19 pandemic.

The sudden outbreak of COVID-19 has dramatically altered the state of the global economy, and the stock market has become more volatile and even fallen sharply as a result of its negative impact, heightening investors' apprehension regarding the correlation between unexpected events and stock market volatility. Additionally, internal and external characteristics coexist in the stock market. Existing research has struggled to extract more effective stock market features during the COVID-19 outbreak using a single time-series neural network model. This paper presents a framework for multitasking learning-based stock market forecasting (COVID-19-MLSF), which can extract the internal and external features of the stock market and their relationships effectively during COVID-19.The innovation comprises three components: designing a new market sentiment index (NMSI) and COVID-19 index to represent the external characteristics of the stock market during the COVID-19 pandemic. Besides, it introduces a multi-task learning framework to extract global and local features of the stock market. Moreover, a temporal convolutional neural network with a multi-scale attention mechanism is designed (MA-TCN) alongside a Multi-View Convolutional-Bidirectional Recurrent Neural Network with Temporal Attention (MVCNN-BiLSTM-Att), adjusting the model to account for the changing status of COVID-19 and its impact on the stock market. Experiments indicate that our model achieves superior performance both in terms of predicting the accuracy of the China CSI 300 Index during the COVID-19 period and in terms of sing market trading.

Analysis of Tuberculosis Disease Spreading Pattern in Muara Enim District using KNN Algorithm

Tuberculosis (TB) is a type of infectious disease caused by Mycobacterium tuberculosis, which not only attacks the lungs, but can also attack the bones, intestines, or glands. During the Covid-19 pandemic, TB cases in Indonesia also increased. TB and Covid-19 had the similar symptoms such as cough, fever, and breathing difficulty, so that TB sufferers must be given serious treatment to avoid Covid-19. In predicting a disease, it is important for health workers to make decisions, thus it is necessary to do an early diagnosis in order to reduce the transmission of TB in the community. There are many algorithm methods used in conducting data analysis, for this study the authors use K-Nearest Neighbor (K-NN) algorithm and Logistic Regression as comparison. Experimental results using available dataset collected from health centers in Muara Enim District of South Sumatra Province show that the K-NN algorithm provides the best accuracy of 89% on dataset with training to testing data ratio of 80%:20%, while the Logistic Regression provides the best accuracy of 96% on 70%:30% ratio. The analysis mechanism discussed in this paper may be considered as tool for the authority to predict and take necessary actions to prevent the TB spreading. © 2022 ACM.

Comparison of Naïve Bayes, C4.5 and K-Nearest Neighbor for Covid-19 Data Classification

Covid-19 is a new virus that appeared in the city of Wuhan in 2019. This virus spreads very quickly even to Indonesia. One effort that can be done to detect the presence of this virus is the PCR and antigen test. Increasing this case resulted in a medical team having difficulty detecting suspects exposed to viruses. This research was conducted to find the best classification algorithm in predicting and classifying status on the suspected Covid-19 both exposed or not exposed. The method used in this study is Naïve Bayes, C4.5 and K-Nearest Neighbor which have very high accuracy using secondary data from the Dumai City Health Agency. From this study it was found that the algorithm C4.5 as the best algorithm in predicting the status of COVID-19 patients, especially in the city of Dumai with an accuracy of 86.54%, recall 71.51%and precision 85.14%. This study has implications for further researchers in choosing an algorithm to predict the COVID-19 case. © 2022 IEEE.

Comparison of Machine Learning Performance for Covid-19 X-ray Image Classification Based on Texture Features

The most prevalent method for early detection of Covid-19 is polymerase chain reaction (PCR). Unfortunately, the quantity of accessible test kits restricts the use of PCR. The development of automatic detection is limited due to the absence of the digital output of PCR data, resulting in an extremely low sensitivity level. Another possibility for Covid-19 detection is based on medical imaging diagnostic. Using digital images offers the opportunity to develop a computer-based system. Image processing mixed with machine learning is the purpose of this study. The comparison of machine learning performance aimed to determine the best classification model. The methods developed for the Covid-19 detection system applied 2-D Haar Wavelet Transform feature extraction and classification methods of Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Decision Tree (DT). Quadratic SVM achieved the best classification results with an accuracy of 86.96%, precision of 94.64%, recall of 86.89%, specificity of 90.00%, and F-score of 89.83%. This study succeeded in comparing three machine learning methods with texture features. © 2022 Institute of Advanced Engineering and Science (IAES).

Implementation of K-Nearest Neighbor to Predict the Chances of COVID-19 Patients' Recovery

Coronavirus Disease 2019 (COVID-19) is a new disease discovered in 2019 in Wuhan, China, and then spread worldwide. Many victims have confirmed varying positive levels of infection based on the patient's immunity. This study aimed to predict the chances of COVID-19 patients' recovery based on the patient's symptoms and conditions. The method used is the K-Nearest Neighbor (KNN) algorithm. KNN produces two classes of predictions: the chance of recovering or the possibility of dying. Based on the experimental results on 496 data from patients who were confirmed positive for COVID-19, KNN predicted the chances of recovery for patients with confirmed COVID-19 with an average accuracy of 88.16%. A prediction system for the chance of recovery for COVID-19 patients is constructed by choosing the best model from five test scenarios based on the given k value. The best model is at a value of k equal to 4, with an accuracy value of 88.8%. © 2022 IEEE.