{"title":"利用 Naive Bayes 将基于过滤器和包装器的特征选择相结合用于心脏病分类","authors":"Siti Roziana Azizah, Rudy Herteno, Andi Farmadi, Dwi Kartini, I. Budiman","doi":"10.25126/jtiik.1067467","DOIUrl":null,"url":null,"abstract":"Penyakit jantung menjadi salah satu penyebab utama kematian bersama dengan penyakit lainnya. Dalam bidang teknologi, data mining dapat digunakan untuk mendiagnosa suatu penyakit yang bersumber dari data rekam medis pasien. Pada klasifikasi dataset medis, Naive Bayes merupakan salah satu metode terbaik yang digunakan. Tujuan dari penelitian ini adalah untuk mengetahui perbandingan hasil akurasi dari Naive Bayes menggunakan beberapa seleksi fitur yaitu Forward Selection, Backward Elimination, kombinasi union hasil seleksi fitur Forwad Selection dan Backward Elimination, Information Gain, Gain Ratio, dan kombinasi union hasil seleksi fitur Information Gain dengan Gain Ratio. Data yang digunakan dalam penelitian ini adalah data penyakit jantung yang didapatkan dari UCI Machine Learning Repository. Dari implementasi pemodelan yang akan dilakukan menghasilkan nilai akurasi tertinggi sebesar 91.80% pada algoritma Naive Bayes dengan kombinasi union hasil seleksi fitur Information Gain dan Gain Ratio menggunakan perbandingan data latih dan data uji 80:20. Sedangkan akurasi Naive Bayes dengan kombinasi union hasil seleksi fitur Forward Selection dan Backward Elimination hanya memiliki nilai akurasi sebesar 83.61% Abstract Heart disease is one of the leading causes of death along with other diseases. In the field of technology, data mining can be used to diagnose a disease sourced from patient medical record data. In the classification of medical datasets, Naive Bayes is one of the best methods used. The purpose of this study is to determine the comparison of the accuracy results of Naive Bayes using several feature selections, namely Forward Selection, Backward Elimination, a combination of union of Forwad Selection and Backward Elimination feature selection results, Information Gain, Gain Ratio, and a combination of union of Information Gain feature selection results with Gain Ratio. The data used in this research is heart disease data obtained from the UCI Machine Learning Repository. From the implementation of modeling that will be carried out, the highest accuracy value is 91.80% in the Naive Bayes algorithm with a combination of union of Information Gain and Gain Ratio feature selection results using a ratio of training data and test data of 80:20. While the accuracy of Naive Bayes with a combination of union selection results of Forward Selection and Backward Elimination features only has an accuracy value of 83.61%.","PeriodicalId":32501,"journal":{"name":"Jurnal Teknologi Informasi dan Ilmu Komputer","volume":" 3","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2023-12-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Kombinasi Seleksi Fitur Berbasis Filter dan Wrapper Menggunakan Naive Bayes pada Klasifikasi Penyakit Jantung\",\"authors\":\"Siti Roziana Azizah, Rudy Herteno, Andi Farmadi, Dwi Kartini, I. Budiman\",\"doi\":\"10.25126/jtiik.1067467\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"Penyakit jantung menjadi salah satu penyebab utama kematian bersama dengan penyakit lainnya. Dalam bidang teknologi, data mining dapat digunakan untuk mendiagnosa suatu penyakit yang bersumber dari data rekam medis pasien. Pada klasifikasi dataset medis, Naive Bayes merupakan salah satu metode terbaik yang digunakan. Tujuan dari penelitian ini adalah untuk mengetahui perbandingan hasil akurasi dari Naive Bayes menggunakan beberapa seleksi fitur yaitu Forward Selection, Backward Elimination, kombinasi union hasil seleksi fitur Forwad Selection dan Backward Elimination, Information Gain, Gain Ratio, dan kombinasi union hasil seleksi fitur Information Gain dengan Gain Ratio. Data yang digunakan dalam penelitian ini adalah data penyakit jantung yang didapatkan dari UCI Machine Learning Repository. Dari implementasi pemodelan yang akan dilakukan menghasilkan nilai akurasi tertinggi sebesar 91.80% pada algoritma Naive Bayes dengan kombinasi union hasil seleksi fitur Information Gain dan Gain Ratio menggunakan perbandingan data latih dan data uji 80:20. Sedangkan akurasi Naive Bayes dengan kombinasi union hasil seleksi fitur Forward Selection dan Backward Elimination hanya memiliki nilai akurasi sebesar 83.61% Abstract Heart disease is one of the leading causes of death along with other diseases. In the field of technology, data mining can be used to diagnose a disease sourced from patient medical record data. In the classification of medical datasets, Naive Bayes is one of the best methods used. The purpose of this study is to determine the comparison of the accuracy results of Naive Bayes using several feature selections, namely Forward Selection, Backward Elimination, a combination of union of Forwad Selection and Backward Elimination feature selection results, Information Gain, Gain Ratio, and a combination of union of Information Gain feature selection results with Gain Ratio. The data used in this research is heart disease data obtained from the UCI Machine Learning Repository. From the implementation of modeling that will be carried out, the highest accuracy value is 91.80% in the Naive Bayes algorithm with a combination of union of Information Gain and Gain Ratio feature selection results using a ratio of training data and test data of 80:20. While the accuracy of Naive Bayes with a combination of union selection results of Forward Selection and Backward Elimination features only has an accuracy value of 83.61%.\",\"PeriodicalId\":32501,\"journal\":{\"name\":\"Jurnal Teknologi Informasi dan Ilmu Komputer\",\"volume\":\" 3\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2023-12-30\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Jurnal Teknologi Informasi dan Ilmu Komputer\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.25126/jtiik.1067467\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Jurnal Teknologi Informasi dan Ilmu Komputer","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.25126/jtiik.1067467","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
数据挖掘是目前最有效的医疗数据挖掘方法之一。数据挖掘技术可以从医疗数据中发现问题并进行诊断。对于医疗数据集,Naive Bayes 是目前最常用的数据挖掘方法。在此数据集中,Naive Bayes 可用于计算前向选择、后向消除、前向选择与后向消除的结合、信息增益、增益比以及信息增益与增益比的结合。从 UCI 机器学习资源库中获取的数据是对计算机的数据分析。通过使用奈维贝叶斯算法,结合信息增益和增益比(Gain Ratio),可以实现高达 91.80% 的最小阈值。在前向选择和后向消除的结合下,奈何贝叶斯算法的准确率达到了 83.61% Abstract Heart disease is one of the leading causes of death along with other diseases.在技术领域,数据挖掘可用于从病人的医疗记录数据中诊断疾病。在医疗数据集的分类中,Naive Bayes 是最好用的方法之一。本研究的目的是确定使用几种特征选择(即前向选择、后向消除、前向选择和后向消除特征选择结果的组合、信息增益、增益比以及信息增益特征选择结果与增益比的组合)的 Naive Bayes 准确性结果的比较。本研究使用的数据是从加州大学洛杉矶分校机器学习资料库中获取的心脏病数据。从即将进行的建模实施来看,在训练数据和测试数据的比例为 80:20 的情况下,采用信息增益和增益比特征选择结果组合的 Naive Bayes 算法的准确率最高,达到 91.80%。而结合了前向选择和后向消除特征选择结果的 Naive Bayes 算法的准确率仅为 83.61%。
Kombinasi Seleksi Fitur Berbasis Filter dan Wrapper Menggunakan Naive Bayes pada Klasifikasi Penyakit Jantung
Penyakit jantung menjadi salah satu penyebab utama kematian bersama dengan penyakit lainnya. Dalam bidang teknologi, data mining dapat digunakan untuk mendiagnosa suatu penyakit yang bersumber dari data rekam medis pasien. Pada klasifikasi dataset medis, Naive Bayes merupakan salah satu metode terbaik yang digunakan. Tujuan dari penelitian ini adalah untuk mengetahui perbandingan hasil akurasi dari Naive Bayes menggunakan beberapa seleksi fitur yaitu Forward Selection, Backward Elimination, kombinasi union hasil seleksi fitur Forwad Selection dan Backward Elimination, Information Gain, Gain Ratio, dan kombinasi union hasil seleksi fitur Information Gain dengan Gain Ratio. Data yang digunakan dalam penelitian ini adalah data penyakit jantung yang didapatkan dari UCI Machine Learning Repository. Dari implementasi pemodelan yang akan dilakukan menghasilkan nilai akurasi tertinggi sebesar 91.80% pada algoritma Naive Bayes dengan kombinasi union hasil seleksi fitur Information Gain dan Gain Ratio menggunakan perbandingan data latih dan data uji 80:20. Sedangkan akurasi Naive Bayes dengan kombinasi union hasil seleksi fitur Forward Selection dan Backward Elimination hanya memiliki nilai akurasi sebesar 83.61% Abstract Heart disease is one of the leading causes of death along with other diseases. In the field of technology, data mining can be used to diagnose a disease sourced from patient medical record data. In the classification of medical datasets, Naive Bayes is one of the best methods used. The purpose of this study is to determine the comparison of the accuracy results of Naive Bayes using several feature selections, namely Forward Selection, Backward Elimination, a combination of union of Forwad Selection and Backward Elimination feature selection results, Information Gain, Gain Ratio, and a combination of union of Information Gain feature selection results with Gain Ratio. The data used in this research is heart disease data obtained from the UCI Machine Learning Repository. From the implementation of modeling that will be carried out, the highest accuracy value is 91.80% in the Naive Bayes algorithm with a combination of union of Information Gain and Gain Ratio feature selection results using a ratio of training data and test data of 80:20. While the accuracy of Naive Bayes with a combination of union selection results of Forward Selection and Backward Elimination features only has an accuracy value of 83.61%.