Klasifikasi Stadium Kanker Payudara Menggunakan Metode Machine Learning dengan Synthetic Minority Oversampling (SMOTE): Studi Kasus Pasien Kanker Payudara Di Rs Onkologi Surabaya

Pratiwi, Dea Restika Augustina (2021) Klasifikasi Stadium Kanker Payudara Menggunakan Metode Machine Learning dengan Synthetic Minority Oversampling (SMOTE): Studi Kasus Pasien Kanker Payudara Di Rs Onkologi Surabaya. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[thumbnail of 06211740000023-Undergraduate_Thesis.pdf] Text
06211740000023-Undergraduate_Thesis.pdf - Accepted Version
Restricted to Repository staff only until 1 October 2023.

Download (1MB) | Request a copy
[thumbnail of 06211740000023-Undergraduate_Thesis.pdf] Text
06211740000023-Undergraduate_Thesis.pdf
Restricted to Repository staff only until 1 October 2023.

Download (1MB) | Request a copy
[thumbnail of 06211740000023-Undergraduate_Thesis.pdf] Text
06211740000023-Undergraduate_Thesis.pdf
Restricted to Repository staff only until 1 October 2023.

Download (1MB) | Request a copy

Abstract

Di Indonesia, kanker merupakan penyebab kematian ketiga setelah jantung dan stroke. Jenis kanker dengan angka kejadian tertinggi pada penduduk wanita adalah kanker payudara. Kanker payudara merupakan tumor ganas yang menyerang jaringan payudara. Klasifikasi stadium kanker payudara penting dilakukan karena menentukan pengobatan yang diberikan pada pasien. Pada penelitian ini akan dilakukan perbandingan dua metode machine learning untuk klasifikasi stadium kanker payudara pasien RS Onkologi Surabaya. Jumlah pasien di setiap stadium mengalami ketimpangan, sehingga dilakukan oversampling dengan SMOTE. Data yang mulanya 294 menjadi 475 data. Data dipartisi menjadi 80% Training dan 20% Testing dengan stratifikasi. Pemilihan parameter terbaik dilakukan dengan menggunakan 10-fold Cross Validation pada data Training. Jumlah neuron sebanyak 8 dan k = 10 merupakan parameter terbaik pada masing-masing model NN dan k-NN. Selanjutnya pemodelan dilakukan dengan metode Neural Network dan k-Nearest Neighbour pada data Training dan Testing. Diperoleh metode Neural Network memiliki nilai AUC yang lebih tinggi dibandingkan k-Nearest Neighbour yaitu sebesar 85,1% sementara k-NN sebesar 83,9%.
================================================================================================
In Indonesia, cancer is the third leading cause of death after heart disease and stroke. The type of cancer with the highest incidence in the female population is breast cancer. Breast cancer is a malignant tumor that attacks the breast tissue. Classification of breast cancer stage is important because it determines the treatment given to the patient. In this study, a comparison of two machine learning methods will be carried out for the classification of breast cancer stages in Surabaya Oncology Hospital patients. The number of patients at each stage experienced inequality, so oversampling was carried out with SMOTE. The data that was originally 294 became 475 data. The data is partitioned into 80% Training and 20% Testing with stratification. The best parameter selection is done by using 10-fold Cross Validation on the Training data. The number of neurons is 8 and k = 10 is the best parameter in each NN and k-NN models. Furthermore, the modeling is carried out using the Neural Network and k-Nearest Neighbor methods on the Training and Testing data. The Neural Network method has a higher AUC value than k-Nearest Neighbor, which is 85.1%, while k-NN is 83.9%.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Area Under Curve, Breast Cancer, k-Nearest Neighbour, Neural Network, Kanker Payudara
Subjects: Q Science > Q Science (General) > Q325.5 Machine learning.
Divisions: Faculty of Science and Data Analytics (SCIENTICS) > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Dea Restika Augustina Pratiwi
Date Deposited: 29 Aug 2021 08:50
Last Modified: 29 Aug 2021 08:50
URI: http://repository.its.ac.id/id/eprint/90776

Actions (login required)

View Item View Item