Predicting COVID-19 Cases in South Korea with All K-Edited Nearest Neighbors Noise Filter and Machine Learning Techniques (original) (raw)
Related papers
Biomedical Signal Processing and Control, 2020
Electromagnetic field (EMF) radiations from mobile phones and cell tower affect brain of humans and other organisms in many ways. Exposure to EMF could lead to neurological changes causing morphological or chemical changes in the brain and other internal organs. Cellular level analysis to measure and identify the effect of mobile radiations is an expensive and long process as it requires preparing the cell suspension for the analysis. This paper presents a novel pilot study to identify changes in brain morphology under EMF exposure considering drosophila melanogaster as a specimen. The brain is automatically segmented, obtaining microscopic images from which discriminatory geometrical features are extracted to identify the effect of EMF exposure. The geometrical features of the microscopic segmented brain image of drosophila are analyzed and found to have discriminatory properties suitable for machine learning. The most prominent discriminatory features were fed to four different classifiers: support vector machine, naïve bayes, artificial neural network and random forest for classification of exposed / non-exposed microscopic image of drosophila brain. Experimental results indicate that all four classifiers provide good classification results up to 94.66 % using discriminatory features selected by feature selection method. The proposed method is a novel approach to identify the effect of EMF exposure automatically and with low time complexity thus providing an efficient image processing framework based on machine learning.
Analysis of COVID-19-Impacted Zone UsingMachine Learning Algorithms
Data Engineering and Communication Technology
Covid-19, first detected at Wuhan in late 2019, has now spread all over the world among many developed and developing countries. As a result of this, World Health Organization (WHO) declared COVID-19 a pandemic on March 11, 2020. Until now, many people have been infected with this coronavirus, some of them are recovering and others causing death. The concern of this paper will be the comparative study of KNN and Naïve Bayes algorithms via the Weka tool's Explorer and Experimenter interfaces, which will tell algorithm is more articulated to be used to evaluate the accuracy of death and recovery of infected COVID-19 patients, so we could estimate that the region will belong to which zone. The COVID-19 dataset to be used in this paper includes details about people who have visited Wuhan during this pandemic or who are from Wuhan and are affected by COVID-19 are fever, cold, cough, breathing difficulties, and many more. The main goal here will be to help users extract valuable data from the dataset and define a predictive algorithm for it. From the results shown, it can be concluded that KNN would demonstrate better precision than Naïve Bayes.
Analyzing of Electromagnetic Exposure from GSM Antennas Using Data Mining Techniques
Nowadays, overall the world the Electromagnetic Fields radiated from telecommunication antennas are main source of electromagnetic pollution in urban, especially when base antennas located adjacent the living areas such as homes, schools, hospitals in city. In this study, electromagnetic pollution measurements were analyzed with 3 different cluster for base stations which is placed at different locations in a city center. K-means algorithm is used for clustering. The data used are obtained from the Report named “Electromagnetic pollution 2012 study in Rize” which is published by the SEMAM (The electromagnetic research center of Sakarya University). Clustering option is determined by the parameters such as the distance from homes to base stations, measurements of electromagnetic field data’s risk status.
SN Computer Science
As the number of COVID-19 cases increases day by day, the situation and livelihood of people throughout the world deteriorates. The goal of this study is to use machine learning models to identify disease and forecast whether or not a person is infected with the virus or another common illness. More articles about COVID-19 will be released starting in 2020, but we still do not have a reliable prediction mechanism to diagnose the disease with 100% accuracy. This comparison is done to see which model is the most effective in detecting and predicting disease. Despite the fact that we have immunizations, we require a best-prediction strategy to assist all humans in surviving. Researchers claimed that the supervised learning method predicts more accurately than the unsupervised learning method in the majority of studies. Supervised learning is the process of mapping inputs to derived outputs using a set of variables and created functions. This will also help us to optimize performance criteria using experience. It is further divided into two categories: classification and regression. According to recent studies, classification models are more accurate than other models.
A Machine Learning technique to analyze and detect Corona Virus
2022
COVID 19 has expanded repeatedly over the whole world, and the number of infected people has been increasing tremendously. COVID 19 has stormed the world in a blink resulting in millions of deaths with economic downfall around the globe. It has triggered a disastrous paradigm shift for the world. Given the unavoidable circumstances, testing for the virus on a rapid daily basis for million people yields the importance of partaking next steps in virus control. The supply chain of traditional Checkup and report time is exorbitant and has the avenue of exceeding the possibility of misreporting. As a result, we have presented Machine learning-based methods for COVID-19 identification. To improve the COVID 19 prediction algorithm, this study indicates the use of exhaustive profiling, SMOTE (Synthetic Minority Oversampling Technique), a classification model, and a deep learning model. This paper goals to provide Machine learning classifier algorithms and Neural Networks with selected attributes to obtain better accuracy and efficacy with a subsequent comparison with different algorithms.
Plant sensitivity and its bio-effects on non-thermal weak radio-frequency electromagnetic fields (RF-EMF) identifying key parameters that affect plant sensitivity that can change/unchange by using big data analytics and machine learning concepts are quite significant. Despite its benefits, there is no single study that adequately covers machine learning concept in Bioelectromagnetics domain yet. This study aims to demonstrate the usefulness of Machine Learning algorithms for predicting the possible damages of electromagnetic radiations from mobile phones and base station on plants and consequently, develops a prediction model of plant sensitivity to RF-EMF. We used raw-data of plant exposure from our previous review study (extracted data from 45 peer-reviewed scientific publications published between 1996-2016 with 169 experimental case studies carried out in the scientific literature) that predicts the potential effects of RF-EMF on plants. We also used values of six different attributes or parameters for this study: frequency, specific absorption rate (SAR), power flux density, electric field strength, exposure time and plant type (species). The results demonstrated that the adaptation of machine learning algorithms (classification and clustering) to predict 1) what conditions will RF-EMF exposure to a plant of a given species may not produce an effect; 2) what frequency and electric field strength values are safer; and 3) which plant species are affected by RF-EMF. Moreover, this paper also illustrates the development of optimal attribute selection protocol to identify key parameters that are highly significant when designing the in-vitro practical standardized experimental protocols. Our analysis also illustrates that Random Forest classification algorithm outperforms with highest classification accuracy by 95.26% (0.084 error) with only 4% of fluctuation among algorithm measured. The results clearly show that using K-Means clustering algorithm, demonstrated that the Pea, Mungbean and Duckweeds plants are more sensitive to RF-EMF (p ≤ 0.0001). The sample size of reported 169 experimental case studies, perhaps low significant in a statistical sense, nonetheless, this analysis still provides useful insight of exploiting Machine Learning in Bioelectromagnetics domain. As a direct outcome of this research, more efficient RF-EMF exposure prediction tools can be developed to improve the quality of epidemiological studies and the long-term experiments using whole organisms.
Classificatin of Brain Tumors by Machine Learning Algorithms (1)
The most prevalent illnesses in the world are skin diseases. Their tough skin texture, presence of hair on the skin, and colour make diagnosis exceedingly challenging. To improve the diagnostic accuracy of many kinds of skin disorders, techniques like machine learning must be developed. The application of machine learning methods in the medical profession is common for diagnosis. In order to decide, these algorithms employ feature values from photos as input. The feature extraction stage, the training stage, and the testing stage are the three steps of the procedure. Utilizing different skin imaging datasets, the technique trains itself using machine learning technologies. The goal of this procedure is to improve the diagnosis of skin diseases. Texture, colour, form, and their combinations are three crucial elements in picture categorisation. In this study, the skin illness is categorised using criteria of colour and texture. The hue of healthy skin differs from that of diseased skin. Using texture attributes in the photos, it is possible to distinguish between smoothness, coarseness, and regularity. In order to successfully diagnose skin illness, these two traits are investigated. In this study, the Hue-Saturation-Value (HSV) characteristics' entropy, variance, and maximum histogram value are employed. These characteristics are used in the Decision Tree (DT) and Support Vector Machine learning algorithms (SVM). Entropy is employed to divide the tree at the first level. Variance is employed at the second level to get leaves for texturing. In colour features, the HSV measure's highest histogram value is utilised to break the tree. The suggested algorithm's performance is evaluated using accuracy.
In this paper, the diagnosis of childhood Atypical Teratoid /Rhabdoid tumor (AT/RT) in magnetic resonance brain images and Hemochromatosis in Computed Tomography (CT) liver images, through the hybridization of particle swarm optimization and firefly (PSO-FF) algorithms for feature selection has been presented. Here, the features are extracted through Non-sub Sampled Contourlet Transform (NSCT) to collect the information in all the directions including the edges from the images, Singular Value Decomposition (SVD) to enhance the image and to get the algebraic details, Gray Level Co-occurrence Matrix (GLCM) method to obtain the statistical textural features from the images. All these features are fused together, then the hybridized meta-heuristics algorithms are applied to extract the salient features from the feature set. The ability of global thinking (gbest) in PSO has been combined with the local search capability of firefly to achieve good results. The Radial Basis Function - Support Vector Machine (RBF-SVM) classifier has been used for the classification of brain and liver disease. The affected part was segmented by using expectation maximization algorithm.
Enhancement of COVID-19 Detection by Unravelling its Structure and Selecting the Optimal Attributes
2021 IEEE Global Communications Conference (GLOBECOM), 2021
According to the current unprecedented pandemic, we realise that we cannot respond to every contagion novel virus as fast as possible, either by vaccination or medication. Therefore, it is paramount for the sustainable development of antiviral urban ecosystems to promote early detection, control, and prevention of an outbreak. The structure of an antivirus-based multi-generational smart-city framework could be crucial to a post-COVID-19 urban environment. Humanitarian efforts in the pandemic's framework deployed novel technological solutions based on the Internet of Things (IoT), Machine Learning, Cloud Computing and Artificial Intelligence (AI). We aim to contribute by improving real-time detection using data mining in collaboration with machine learning techniques through our research work. Initially, for detection, we propose an innovative system that could detect in real-time virus propagation based on the density of the airborne COVID-19 molecules-the proposal based on the detection through the isothermal amplification RT-Lamp [1]. We also propose real-time detection by spark-induced plasma spectroscopy during the internal airborne transmission process [17]. The novelty of this research work, called characteristic subset selection, is based on identifying irrelevant data. By deducting the unrelated information dimension, machine learning algorithms would operate more efficiently. Therefore, it optimises data mining and classification in high-dimensional medical data analysis, particularly in effectively detecting COVID-19. It can play an essential role in providing timely detection with critical attributes and high accuracy. We elaborate the teaching-learning method optimisation to achieve the optimal set of features for the detection.
Journal of Experimental & Theoretical Artificial Intelligence, 2016
Magnetic resonance imaging segmentation refers to a process of assigning labels to set of pixels or multiple regions. It plays a major role in the field of biomedical applications as it is widely used by the radiologists to segment the medical images input into meaningful regions. In recent years, various brain tumour detection techniques are presented in the literature. The entire segmentation process of our proposed work comprises three phases: threshold generation with dynamic modified region growing phase, texture feature generation phase and region merging phase. by dynamically changing two thresholds in the modified region growing approach, the first phase of the given input image can be performed as dynamic modified region growing process, in which the optimisation algorithm, firefly algorithm help to optimise the two thresholds in modified region growing. After obtaining the region growth segmented image using modified region growing, the edges can be detected with edge detection algorithm. In the second phase, the texture feature can be extracted using entropy-based operation from the input image. In region merging phase, the results obtained from the texture feature-generation phase are combined with the results of dynamic modified region growing phase and similar regions are merged using a distance comparison between regions. After identifying the abnormal tissues, the classification can be done by hybrid kernel-based SVM (Support Vector Machine). The performance analysis of the proposed method will be carried by K-cross fold validation method. The proposed method will be implemented in MATLAB with various images.