Perbandingan tingkat akurasi algoritma support vector machine dan k-nearest neighbor dalam mendeteksi penyakit serangan jantung

Daniswara, Erick (2023) Perbandingan tingkat akurasi algoritma support vector machine dan k-nearest neighbor dalam mendeteksi penyakit serangan jantung. Bachelor thesis, Universitas Pelita Harapan.

Preview	Text (Title) Title.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (164kB) | Preview
Preview	Text (Abstract) Abstract.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (759kB) | Preview
Preview	Text (ToC) ToC.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (612kB) | Preview
Preview	Text (Chapter1) Chapter1.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (778kB) | Preview
	Text (Chapter2) Chapter2.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)
	Text (Chapter3) Chapter3.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB)
	Text (Chapter4) Chapter4.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (3MB)
	Text (Chapter5) Chapter5.pdf Restricted to Registered users only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (404kB)
Preview	Text (Bibliography) Bibliography.pdf Available under License Creative Commons Attribution Non-commercial Share Alike. Download (1MB) | Preview
	Text (Appendices) Appendices.pdf Restricted to Repository staff only Available under License Creative Commons Attribution Non-commercial Share Alike. Download (2MB)

Abstract

Serangan jantung merupakan penyakit mematikan dengan angka kematian yang mencapai 36% pada tahun 2020. Ini harus diberikan perhatian yang lebih dalam mengantisipasi resiko penyakit serangan jantung. Kontribusi dalam bidang teknologi yang dapat diberikan adalah informasi terkait resiko serangan jantung. Pada penelitian sebelumnya, ada peneliti yang menggunakan algoritma Naïve Bayes dengan akurasi sebesar 84%, dan Logistic Regression sebesar 88%. Oleh karena itu, dalam penelitian ini memiliki ketertarikan dalam algoritma Support Vector Machine (SVM) dan K-Nearest Neighbor (KNN) untuk melakukan perbandingan nilai akurasi antara kedua algoritma tersebut dengan evaluasi model confusion matrix. Sumber data pasien serangan jantung yang diperoleh berasal dari Kaggle. Data tersebut memiliki 303 data dengan 14 kolom atribut yang akan dibagi menjadi 70% data training dan 30% data testing dari keseluruhan data. Softwareyang digunakan dalam penelitian ini yaitu Orange Data Mining untuk membangun Support Vector Machine (SVM) dan K-Nearest Neighbor (KNN). Hasil dari penelitian yaitu akurasi Support Vector Machine (SVM) sebesar 85.6%, sedangkan K-Nearest Neighbor (KNN) sebesar 81.1%. Pada confusion matrix tingkatkesalahan yang dihasilkan algoritma Support Vector Machine lebih kecildibandingkan K-Nearest Neighbor (KNN). Kesimpulan yang dapat diberikan yaitualgoritma Support Vector Machine (SVM) lebih unggul daripada K-NearestNeighbor (KNN) dalam memprediksi penyakit serangan jantung. / Heart attack is a deadly disease with a mortality rate that reached 36% in 2020. This should be given more attention in anticipating the risk of heart attack disease. The contribution in the field of technology that can be given is information related to the risk of heart attack. In previous studies, there were researchers who used the Naïve Bayes algorithm with an accuracy of 84%, and Logistic Regression of 88%. Therefore, this research has an interest in the Support Vector Machine (SVM) and K-Nearest Neighbor (KNN) algorithms to compare the accuracy value between the two algorithms by evaluating the confusion matrix model. The source of heart attack patient data obtained comes from Kaggle. The data has 303 data with 14 attribute columns which will be divided into 70% training data and 30% testing data from the entire data. The software used in this research is Orange Data Mining to build Support Vector Machine (SVM) and K-Nearest Neighbor (KNN). The results of the research are the accuracy of Support Vector Machine (SVM) of 85.6%, while K-Nearest Neighbor (KNN) of 81.1%. In the confusion matrix, the error rate produced by the Support Vector Machine algorithm is smaller than the K-Nearest Neighbor (KNN). The conclusion that can be given is that the SupportVector Machine (SVM) algorithm is superior to K-Nearest Neighbor (KNN) inpredicting heart attack disease.

Actions (login required)

View Item