@article{ff89f8a9eb064b0f9092c6e458747dd7,
title = "Classification of thalassemia data using K-nearest neighbor and Na{\"i}ve Bayes",
abstract = "Thalassemia is an abnormal blood disorder. Research on thalassemia has been done a lot, such as testing on beta-thalassemia data using a decision tree, K-Nearest Neighbor, and Multi-Layer Perceptron classifier. In this study, we will discuss the comparative performance of classification of thalassemia data using K-Nearest Neighbor and Na{\"i}ve Bayes which are very popular in the field of classification. The author uses thalassemia data from Harapan Kita Hospital, West Jakarta. While the training data used ranges from 10 to 90, the results of the Na{\"i}ve Bayes classification are higher than the classification results using K-Nearest Neighbor with an average of 99.78% with an average running time of 0.06 seconds, while the KNN is 97.14% with the average running time 0.081 seconds.",
keywords = "Classification, K-Nearest Neighbor, Na{\"i}ve Bayes, Thalassemia",
author = "Titin Siswantining and Annisa Kamalia and Zuherman Rustam and Fajar Subroto and Semendawai, {Aditya Suryansyah}",
note = "Funding Information: The results that will be displayed are the average of 10 times running program results. The following table: In the tests that have been carried out, it was found that the highest average value of accuracy in KNN was obtained when the parameter K = 1. So, here is shown the experimental results on thalassemia data using KNN with K = 1. It can be seen that the resulting accuracy varies from 70% to 97%. The biggest accuracy value is when 90% training data (10% testing data) is equal to 97.14% with running time 0.081099 seconds. It can be seen that the accuracy of using Na{\"i}ve Bayes is quite good with a range of 98.5% to 99.7%. The biggest accuracy value is when the training data is 40% (60% data testing) with running time 0.55451 seconds 5. Conclusion In this study, we have classified thalassemia data from Harapan Kita hospital using KNN (K = 1) and Na{\"i}ve Bayes classifier. When the training data used ranges from 10% to 90%, the accuracy value using Na{\"i}ve Bayes is 99.78% with running time of 0.06 seconds, while the KNN accuracy value is 97.14% with running time 0.08 seconds. Thus, it can be concluded that Na{\"i}ve Bayes is better than KNN in terms of classifying thalassemia data. 6. Future Work Furthermore, this study can be used as a basis for thalassemia research using KNN classifier and Na{\"i}ve Bayes classifier with development or using other methods. Acknowledgement This research was supported financially by the Indonesia Ministry of Research and Higher Education, with PDUPT 2018 research grant scheme ID number 389/UN2.R3.1/HKP05.00/2018. References [1] World Health Organization (WHO). 2013. Thalassemia. Funding Information: This research was supported financially by the Indonesia Ministry of Research and Higher Education, with PDUPT 2018 research grant scheme ID number 389/UN2.R3.1/HKP05.00/2018. Publisher Copyright: {\textcopyright} 2019 SERSC.",
year = "2019",
month = oct,
day = "8",
language = "English",
volume = "28",
pages = "15--19",
journal = "International Journal of Advanced Science and Technology",
issn = "2005-4238",
publisher = "Science and Engineering Research Support Society",
number = "8 Special Issue",
}