Perbandingan Performa Antara Algoritma Naïve Bayes Dan K-Nearest Neighbour Pada Klasifikasi Kanker Payudara

Annisa, Nugraheni (2021) Perbandingan Performa Antara Algoritma Naïve Bayes Dan K-Nearest Neighbour Pada Klasifikasi Kanker Payudara. Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (329kB)
	Text ABSTRACT.pdf Download (32kB)
	Text ABSTRAK.pdf Download (33kB)
	Text BAB I.pdf Download (1MB)
	Text BAB II.pdf Download (1MB)
	Text BAB III.pdf Download (1MB)
	Text BAB IV.pdf Restricted to Registered users only Download (1MB) | Request a copy
	Text BAB V.pdf Download (1MB)
	Text DAFTAR PUSTAKA.pdf Download (1MB)
	Text Lampiran.pdf Restricted to Registered users only Download (1MB) | Request a copy

Abstract

Breast cancer occurs when cells in breast tissue begin to have growths that cannot be controlled and can interfere with healthy tissue. Breast cancer is the disease with the highest number of patients in Indonesia and ranks 2nd as a disease that causes cancer death after lung cancer in the first place in Indonesia. The number of cases of breast cancer in Indonesia needs to be considered with preventive measures and early detection of breast cancer. One way to detect breast cancer is to classify breast cancer patients with healthy people. Based on previous research, the Naïve Bayes and K-Nearest Neighbour classification algorithms have the highest accuracy compared to other algorithms and are considered suitable for the classification of numerical and nominal type data. The data used in this study is the data on routine blood checks, namely the Breast Cancer Coimbra dataset in 2018 taken from the UCI Machine Learning Repository which is numerical data. In this study the performance of Naïve Bayes and K-NN algorithms was measured by confusion matrix (accuracy, precision, and recall) and ROC-AUC curve to determine the most precise algorithm in predicting breast cancer. In testing the algorithm, using several testing scenarios, namely, testing data before and after normalization, testing models based on comparison of training data and data testing, testing models based on K-NN values, and testing models based on the selection of the strongest attributes with Pearson correlation tests. The results of this study showed that Naïve Bayes' algorithm had the highest average accuracy of 69.12%, healthy precision 64.90%, painful precision 83%, healthy recall 88%, sick recall 61.11% and AUC 0.82 which belonged to the good classification category. As for the highest average results of the K-Nearest Neighbour algorithm is 76.83% accuracy, healthy precision 76%, painful precision 80.21%, 74.18% of healthy recalls, sick recalls 80.81% and AUC 0.91 which belongs to the category of excellent classification. Keywords: Breast Cancer, Performance Test, Naïve Bayes, K-Nearest Neighbor, and Confusion Matrix

Actions (login required)

View Item