Klasifikasi Microarray “Prostate Cancer” Menggunakan Metode Fuzzy Support Vector Machine (Fsvm)-Genetic Algorithm

PRATIWI, CICILIA AJENG (2018) Klasifikasi Microarray “Prostate Cancer” Menggunakan Metode Fuzzy Support Vector Machine (Fsvm)-Genetic Algorithm. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
06211645000019_Undergraduate_Theses..pdf - Accepted Version

Download (3MB) | Preview

Abstract

Salah satu jenis kanker yang menjadi penyebab terbanyak kematian pada populasi pria adalah kanker prostat. Penyakit ini hanya terdapat pada pria karena pada wanita tidak memiliki ke-lenjar prostat. Secara global, kanker prostat menduduki urutan keempat kanker yang paling sering ditemukan pada manusia sete-lah kanker payudara, paru dan kolorektum untuk angka kejadian kanker pada pria, kanker prostat menduduki urutan ke-2. Pada umumnya penderita baru mengetahui penyakit tersebut sudah me-masuki stadium lanjut. Terlambatnya penanganan pada penderita prostate bisa berakibat fatal bahkan dapat menye-babkan kema-tian. Oleh karena itu, penyakit kanker prostat sangat penting un-tuk didiagnosis sedini mungkin sebelum penyebaran sel kanker ke organ internal. Pada perkembangan saat ini, terdapat teknologi microarray yang memiliki pengaruh besar dalam menentukan gen informatif menyebabkan kanker. Ekspresi gen yang terdapat pada data microarray “prostat” dapat digunakan untuk mengklasifika-sikan pasien yang mengalami tumor prostat dan normal. Klasifi-kasi Fuzzy Support Vector Machine (FSVM) dengan seleksi Fast Correlation Based Filter (FCBF) tanpa optimasi genetic algorithm menghasilkan nilai akurasi lebih tinggi dibandingkan tanpa seleksi. Selain itu, diperoleh nilai akurasi klasifikasi FSVM menggunakan seleksi dan optimasi genetic algorithm lebih tinggi dibandingkan tanpa seleksi ==================================================================================================== One type of cancer that causes the most deaths in the male population is Prostate Cancer. This disease is only found on men because women do not have prostate gland. Globally, Prostate cancer ranked 4th as the most common cancer found in humans after breast, lung and colorectal cancer, while the number of cancers in men, prostate cancer ranked 2nd. Generally, the patients start to know and feel this kind of disease when it entered to the serious level. Late handling in prostate patients can be fatal and can even cause death. Therefore, it is a must to diagnose prostate cancer as early as possible before it’s enlarged to internal organs. In the current development, there are microarray technologies that have major influence in determining the informative genes that causes cancer. This study use microarray “prostate cancer” data that have been done by Dinesh Singh with his friends in 2002. Gene expression contained in “prostate” microarray data can be used to classify patients with prostate and normal (without prostate). Fuzzy Support Vector Machine (FSVM) classification with Fast Correlation Based Filter (FCBF) selection without genetic algorithm optimization resulting in more accuracy higher than without the selection, and also the accuracy of FSVM classification with selection and genetic algorithm optimization is higher than without selection.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: FCBF, feature selection, fuzzy support vector machine, genetic algorithm, microarray.
Subjects: H Social Sciences > HA Statistics
H Social Sciences > HA Statistics > HA29 Theory and method of social science statistics
H Social Sciences > HA Statistics > HA31.7 Estimation
H Social Sciences > HM Sociology
Q Science > Q Science (General)
Q Science > Q Science (General) > Q325 GMDH algorithms.
Q Science > QH Biology > QH301 Biology
Divisions: Faculty of Mathematics and Science > Statistics > 49201-(S1) Undergraduate Thesis
Depositing User: Cicilia Ajeng Pratiwi
Date Deposited: 22 Jul 2021 21:12
Last Modified: 22 Jul 2021 21:12
URI: https://repository.its.ac.id/id/eprint/57561

Actions (login required)

View Item View Item