Image of KLASIFIKASI CITRA DERMOSKOPI KANKER KULIT DENGAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN)

Text

KLASIFIKASI CITRA DERMOSKOPI KANKER KULIT DENGAN MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK (CNN)

Penilaian

0,0

dari 5
Penilaian anda saat ini :  

A Dermoscopy image is one way to visually diagnose skin cancer. Dermoscopy allows for the visualization of subsurface skin structures in the epidermis, these structures are usually not visible to the naked eye. This makes the Demorscopy image very suitable to be a source of classification data. In this research, the software is built to classify skin cancer using the Convolutional Neural Network method. The Convolutional Neural Network (CNN) architectural model created in this research will be compared with the 3 other CNN models that have been previously available, Mobile Net, ResNet50 V2, and VGG16. The model trained using the HAM10000 dataset which contains 7 types of skin cancer, with 27.385 training data, 6847 validation data dan 8559 test data. The Mobile Net model has the highest performance at 95% accuracy, 95% sensitivity, 95% f1-score, 96% precision, and 99% specificity, followed by ResNet50 V2 which has 93% accuracy , 93% precision, 93% sensitivity, 93% f1-score ,and 99% specificity, then VGG16 which has 92% accuracy, 92% sensitivity, 92% f1-score, 93% precision, and 99% specificity, and finally the CNN model which has 87% accuracy, 87% precision, 87% sensitivity, 87% f1-score ,and 98% specificity.

Availability
Inventory Code Barcode Call Number Location Status
2207002482T75120T751202022Central Library (Referens)Available but not for loan - Not for Loan
Detail Information
Series Title
-
Call Number
T751202022
Publisher
Inderalaya : Jurusan Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya.,
Collation
xvii, 129 hlm.; ilus.; 29 cm
Language
Indonesia
ISBN/ISSN
-
Classification
006.307
Content Type
-
Media Type
-
Carrier Type
-
Edition
-
Subject(s)
Sistem Pakar
Jurusan Teknik Informatika
Specific Detail Info
-
Statement of Responsibility
Other version/related

No other version available

File Attachment
Comments
You must be logged in to post a comment