SISTEM AUTENTIKASI GERAKAN TETIKUS MENGGUNAKAN PRINCIPAL COMPONENT ANALYSIS DAN KERNELIZED ONECLASS SUPPORT VECTOR MACHINE

Fauzi Nur Firman, - (2019) SISTEM AUTENTIKASI GERAKAN TETIKUS MENGGUNAKAN PRINCIPAL COMPONENT ANALYSIS DAN KERNELIZED ONECLASS SUPPORT VECTOR MACHINE. S1 thesis, Universitas Pendidikan Indonesia.

Baca Full Text klik disini
[img] Text
S_KOM_1407206_Title.pdf
Download (145kB)
[img] Text
S_KOM_1407206_Abstract.pdf
Download (82kB)
[img] Text
S_KOM_1407206_Table_of_Content.pdf
Download (87kB)
[img] Text
S_KOM_1407206_Chapter1.pdf
Download (134kB)
[img] Text
S_KOM_1407206_Chapter2.pdf
Restricted to Staf Perpustakaan
Download (977kB)
[img] Text
S_KOM_1407206_Chapter3.pdf
Download (516kB)
[img] Text
S_KOM_1407206_Chapter4.pdf
Restricted to Staf Perpustakaan
Download (443kB)
[img] Text
S_KOM_1407206_Chapter5.pdf
Download (51kB)
[img] Text
S_KOM_1407206_Bibliography.pdf
Download (198kB)
[img] Text
S_KOM_1407206_Appendix.pdf
Restricted to Staf Perpustakaan
Download (1MB)
Official URL: http://repository.upi.edu

Abstract

Autentikasi merupakan bagian dari sistem keamanan, yang bertujuan untuk mengenali pengguna apakah dia berhak atau tidak terhadap sebuah sistem. Autentikasi Biometrik mouse menjadi salah cara dalam mengenali pengguna. Saat fitur mouse yang diekstrak cukup banyak maka diperlukan kombinasi fungsi kernel dan support vector machine. Fungsi kernel sebagai pengubah non-linier memberikan beban komputasi yang tinggi. Tetapi terjadi permasalahan bila data yang diolah memiliki dimensi yang tinggi. Akhirnya pengurangan fungsi kernel pada reduksi dimensi fitur dilakukan. Maka dari itu diterapkan principal component analysis sebagai metode reduksi dimensi tanpa fungsi kernel untuk mengurangi dimensi data. Hasil eksperimen menunjukkan tingkat FAR, FRR, dan HTER sebesar 19.65% , 28.13%, dan 23.89%. Waktu training dan autentikasi yang dihasilkan selama 7.15 detik dan 0.17 detik.----------Authentication is a part of a security system, which aims to recognize the user whether he is entitled or not to a system. Authentication Mouse biometrics is one way to recognize users. When the mouse features are extracted quite a lot, we need a combination of kernel functions and support vector machine. The kernel function as a non-linear modifier provides a high computational load. But problems occur if the processed data has a high dimension. Finally the reduction of kernel functions in the reduction of feature dimensions is done. Therefore principal component analysis is applied as a dimensional reduction method without kernel functions to reduce data dimensions. The experimental results showed the levels of FAR, FRR, and HTER were 19.65%, 28.13%, and 23.89% respectively. Training and authentication times are generated for 7.15 seconds and 0.17 seconds.

Item Type: Thesis (S1)
Additional Information: No. Panggil: S KOM FAU s-2019; Pembimbing: I. Munir, II. Erna Piantari; NIM: 1407206
Uncontrolled Keywords: Autentikasi, Biometrik, Pergerakan Mouse, Fungsi Kernel, Principal Component Analysis, Support Vector Machine.
Subjects: L Education > L Education (General)
Q Science > QA Mathematics > QA76 Computer software
Divisions: Fakultas Pendidikan Matematika dan Ilmu Pengetahuan Alam > Program Studi Ilmu Komputer
Depositing User: Santi Santika
Date Deposited: 04 Jun 2020 02:22
Last Modified: 04 Jun 2020 02:22
URI: http://repository.upi.edu/id/eprint/49047

Actions (login required)

View Item View Item