Skripsi

KLASIFIKASI EMOSI MANUSIA MENGGUNAKAN CONVOLUTIONAL NEURAL NETWORK PADA DATASET ELECTROENCEPHALOGRAM

---

Advances in signal processing technology and machine learning have created opportunities to leverage electroencephalogram (EEG) signals for more advanced human emotion recognition. This study aims to develop a human emotion classification model based on a Convolutional Neural Network (CNN) using the SEED-IV dataset, focusing on four emotion classes: happy, sad, fear, and neutral. The data consist of 1,080 EEG signal segments, with 270 segments for each emotion class, extracted using differential entropy features across five frequency bands and mapped into a three-dimensional array representation. The data were then split into training and testing sets with an 80:20 ratio. Several 3D CNN architecture scenarios were evaluated by varying the number of convolutional layers, the number of neurons in the fully connected layer, the number of training epochs, and the optimizer type. Performance evaluation was conducted using precision, recall, and F1-score derived from the confusion matrix. The results show that the best model achieved an accuracy of 76%. These findings indicate that a 3D CNN architecture using a topographic representation of EEG signals has potential as a foundation for developing human emotion recognition systems.

---

Availability

Detail Information

Series Title

-

Call Number

T1888242025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xiii, 98 hlm.; ilus.; tab, 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Convolutional Neural Network

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment