CD Skripsi

Penerapan Arsitektur Efficientnet-B0 Dalam Identifikasi Alfabet Sistem Isyarat Bahasa Indonesia Menggunakan Algoritma CNN

---

The Indonesian Sign Language System (SIBI) is the official sign language used by the deaf in Indonesia for communication. The lack of knowledge about sign language makes it difficult to establish communication between the deaf and the general public. This research aims to facilitate communication between the deaf and the general public, or vice versa, by utilizing deep learning technology in artificial intelligence to process image data. The Convolutional Neural Network (CNN) algorithm is used by applying the EfficientNet-B0 architecture to develop the SIBI Alphabet Identification model through the stages of data collection, pre-processing, model development, training and evaluation, as well as web development. With a dataset of 1.560 images and 26 classes, letters A to Z, the model achieved a training accuracy of 97.56%, a validation accuracy of 89.10%, and a testing accuracy of 96%. Based on the metric scores in the classification report, most classes can be classified well except for the letters E and S due to lower metric scores. The EfficientNet-B0 model was successfully implemented in the development of the SIBI Alphabet Identification web application. This research is expected to serve as a foundational basis for developing subsequent models and practical applications.

Keywords: Convolutional Neural Network, Deaf, Deep Learning, EfficientNet-B0, Indonesian Sign Language System

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2103112035

Penerbit

Pekanbaru : Universitas Riau FMIPA Sistem Informasi., 2025

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2103112035

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

SISTEM INFORMASI

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Mutia

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar