Perbandingan Performa Naïve Bayes & Knearest Neighbor dalam Mengklasifikasi Status Pasien Coronavirus Disease (Covid-19)
Total View This Week0
Total View This Week0
Institusion
Institut Teknologi Telkom Purwokerto
Author
Danang, Wicaksana
Subject
T Technology (General)
Datestamp
2021-09-24 07:24:44
Abstract :
There is no priority for the treatment of COVID-19 patients based on Age, Gender, and Nationality attributes, therefore a comparison of the performance of 2 algorithms, namely Naive Bayes and K-Nearest Neighbor, is to classify the status of COVID-19 patients based on Age, Gender, and Nationality attributes. The dataset used to conduct classification is the COVID-19 dataset for Indonesia and India which is accessed from the Kaggle website,, namely https://www.kaggle.com/ardisragen/indonesia-coronavirus-cases?select=pati ent.csv for the Indonesian dataset and https://www.kaggle.com/sudalairajkumar/covid19-in-india/version/204?select =IndividualDetails.csv for the Indian dataset. The method used in this research is Naïve Bayes and K-nearest Neighbor. By using the confusion matrix measurement method, the results obtained from the K-Nearest Neighbor algorithm have a better accuracy rate of 90.416%, while Naïve Bayes has an accuracy rate of 85%. The precision obtained by Naïve Bayes is 24% for released class, 0.93% for an isolated class, and 0% for the deceased class. The recall obtained by Naïve Bayes is 35% for a released class, 0.91% for isolated class, and 0% for a deceased class. The precision obtained by a K-Nearest Neighbor is 0% for released class, 91% for isolated class, and 0% for deceased class. The recall obtained by a K-Nearest Neighbor is 0% for released class, 100% for isolated class, and 0% for deceased class. Keyword: Coronavirus, COVID-19, Classification, Naïve Bayes, K-Nearest Neighbor
There is no priority for the treatment of COVID-19 patients based on Age, Gender, and Nationality attributes, therefore a comparison of the performance of 2 algorithms, namely Naive Bayes and K-Nearest Neighbor, is to classify the status of COVID-19 patients based on Age, Gender, and Nationality attributes. The dataset used to conduct classification is the COVID-19 dataset for Indonesia and India which is accessed from the Kaggle website,, namely https://www.kaggle.com/ardisragen/indonesia-coronavirus-cases?select=pati ent.csv for the Indonesian dataset and https://www.kaggle.com/sudalairajkumar/covid19-in-india/version/204?select =IndividualDetails.csv for the Indian dataset. The method used in this research is Naïve Bayes and K-nearest Neighbor. By using the confusion matrix measurement method, the results obtained from the K-Nearest Neighbor algorithm have a better accuracy rate of 90.416%, while Naïve Bayes has an accuracy rate of 85%. The precision obtained by Naïve Bayes is 24% for released class, 0.93% for an isolated class, and 0% for the deceased class. The recall obtained by Naïve Bayes is 35% for a released class, 0.91% for isolated class, and 0% for a deceased class. The precision obtained by a K-Nearest Neighbor is 0% for released class, 91% for isolated class, and 0% for deceased class. The recall obtained by a K-Nearest Neighbor is 0% for released class, 100% for isolated class, and 0% for deceased class. Keyword: Coronavirus, COVID-19, Classification, Naïve Bayes, K-Nearest Neighbor
Download
book
BibTex
Latex, Jabref
cloud_download
Original Resource
url resource Institution
