Belangi, Hera Amelia Putri (2023) Komparasi Performa Algoritma Convolutional Neural Network (CNN) dan Support Vector Machine (SVM) pada Studi Kasus Klasifikasi Penyakit Alzheimer berbasis Data MRI. Undergraduate thesis, UPN Veteran Jawa Timur.
|
Text (Cover)
19081010145_cover.pdf Download (1MB) | Preview |
|
|
Text (Bab 1)
19081010145_bab1.pdf Download (155kB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 2)
19081010145_bab2.pdf Restricted to Registered users only until 25 September 2025. Download (17MB) | Request a copy |
|
|
Text (Bab 3)
19081010145_bab3.pdf Restricted to Registered users only until 25 September 2025. Download (1MB) | Request a copy |
|
|
Text (Bab 4)
19081010145_bab4.pdf Restricted to Registered users only until 25 September 2025. Download (12MB) | Request a copy |
|
|
Text (Bab 5)
19081010145_bab5.pdf Download (138kB) | Preview |
|
|
Text (Daftar Pustaka)
19081010145_daftarpustaka.pdf Download (114kB) | Preview |
Abstract
Alzheimer's disease is a neurodegenerative disorder that leads to a decline in memory, cognitive abilities, speech capabilities, and changes in behavior in individuals. Currently, the diagnosis of Alzheimer's disease relies on clinical signs and neuropsychological tests, but these methods lack sufficient accuracy. Therefore, the use of MRI (Magnetic Resonance Imaging) has emerged as a promising alternative to enhance the accuracy of Alzheimer's disease diagnosis. This study compares the performance of two classification algorithms, namely CNN (Convolutional Neural Network) and SVM (Support Vector Machine), based on MRI data. The dataset consists of four classes, including Non Demented, Very Mild Demented, Mild Demented, and Moderate Demented. The results indicate that the CNN method using transfer learning with the EfficientNetB architecture performs better compared to the SVM method with CNN feature extraction in classifying Alzheimer's disease based on MRI data. The accuracy achieved by the CNN method using transfer learning with the best architecture, EfficientNetB1, reaches 98.25%, whereas the accuracy of the SVM method with CNN feature extraction, using the RBF Kernel, reaches a maximum of 79.75%. Thus, the CNN method using transfer learning is considered the more effective algorithm with the best performance in Alzheimer's disease classification.
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Hera Amelia Putri Belangi | ||||||||||||
| Date Deposited: | 25 Sep 2023 03:40 | ||||||||||||
| Last Modified: | 25 Sep 2023 03:40 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/18147 |
Actions (login required)
 |
View Item |