Prosiding Seminar Nasional Ilmu Komputer Universitas Semarang 2021
Optimasi Algoritma Naïve Bayes dengan Menerapkan Discretization dan Particle Swarm Optimization untuk Diagnosis Breast Cancer Disease

Prosiding Seminar Nasional Ilmu Komputer Universitas Semarang 2021
Optimasi Algoritma Naïve Bayes dengan Menerapkan Discretization dan Particle Swarm Optimization untuk Diagnosis Breast Cancer Disease

Algoritma Naïve Bayes, Discretization, Particle Swarm Optimization, Breast Cancer Coimbra Disease
Naïve Bayes Algorithm, Discretization, Particle Swarm Optimization, Breast Cancer Coimbra Disease

Data mining adalah proses mempekerjakan satu atau lebih machine learning untuk menganalisis dan mengekstraksi pengetahuan secara otomatis dengan pendekatan matematis. Naïve Bayes merupakan Algoritma yang sering digunakan dalam data mining. Data Breast Cancer Coimbra Disease yang digunakan diambil dari UCI Machine Learning Repository. Discretization digunakan untuk melakukan preprocessing data pada penelitian. Untuk meningkatkan akurasi, Algoritma Naïve Bayes dapat digabungkan dengan feature selection Particle Swarm Optimization. Penelitian bertujuan untuk mengetahui bagaimana penerapan Algoritma Naïve Bayes digabungkan dengan Discretizationdan dan Particle Swarm Optimization pada diagnosis Breast Cancer Coimbra Disease dan mengetahui peningkatan akurasinya. Berdasarkan hasil sepuluh k-fold cross validation hasil akurasi yang didapat diperoleh akurasi tertinggi fold ke 2 sebesar 91.67%, akurasi terendah pada fold ke 3 sebesar 41.67%, dan rata-rata akurasi sebesar 66.59%. Penerapan Discretization dan Particle Swarm Optimization mampu meningkatkan akurasi dari diagnosis Breast Cancer Coimbra Disease pada akurasi terbaik sebesar 8.34%, akurasi terendah sebesar 23.49%, dan rata-rata akurasi sebesar 11.82%.
Data mining is the process of employing one or more machine learning to analyze and extract knowledge automatically with a mathematical approach. Naïve Bayes is an algorithm that is often used in data mining. Data on Breast Cancer Coimbra Disease used were taken from the UCI Machine Learning Repository. Discretization is used to preprocess data in research. To improve accuracy, the Naïve Bayes Algorithm can be combined with the Particle Swarm Optimization feature selection. This study aims to determine how the application of the Naïve Bayes Algorithm is combined with Discretization and Particle Swarm Optimization in the diagnosis of Breast Cancer Coimbra Disease and to find out the increase in its accuracy. Based on the results of ten k-fold cross validation, the accuracy results obtained were the highest accuracy for the second fold of 91.67%, the lowest accuracy on the 3rd fold was 41.67%, and the average accuracy was 66.59%. The application of Discretization and Particle Swarm Optimization was able to increase the accuracy of the diagnosis of Breast Cancer Coimbra Disease with the best accuracy of 8.34%, the lowest accuracy of 23.49%, and the average accuracy of 11.82%.

Fiscall Saktiyana Aditama, Endang Sugiharti

Universitas Semarang

13 Oktober 2021

Sri Wahyuni

ISSN: 2614-1205

PDF

Indonesian

Text

Prosiding Seminar Nasional Ilmu Komputer Universitas Semarang 2021