Klasifikasi Ekspresi Siswa dalam Pembelajaran Daring Berdasarkan Deteksi Wajah dengan Menggunakan Pendekatan Deep Learning

Fatimah, Clarissa (2023) Klasifikasi Ekspresi Siswa dalam Pembelajaran Daring Berdasarkan Deteksi Wajah dengan Menggunakan Pendekatan Deep Learning. Other thesis, Institut Teknologi Sepuluh Nopember.

Text 05311940000012-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (3MB) | Request a copy

Abstract

Pandemi COVID19 mendorong para siswa dan tenaga pendidik untuk melihat bahwa perkembangan teknologi telah beradaptasi dengan cepat, salah satu contohnya adalah penerapan pembelajaran daring. Pada studi yang dilakukan, pergeseran paradigma pada pembelajaran daring telah mendapatkan revelansi dalam pendidikan universitas dan diperkirakan dapat bertahan keberlangsungannya bahkan saat pasca pandemik. Pada penelitian ini akan dilakukan pemanfaatan teknologi pengenalan wajah Facial Expression Recognition (FER) dimana digunakan untuk mengenali atau mengidentifikasi ketertarikan siswa terhadap materi yang disampaikan berdasarkan ekspresi wajah. Maka dari itu, dengan adanya penelitian ini akan dapat membantu tenaga pendidik dalam mengetahui ekspresi siswa dalam pembelajaran daring. Pada penelitian ini akan dilakukan pembuatan model klasifikasi ekspresi wajah dengan menggunakan arsitektur VGG16 pada dataset FER2013 dimana terdapat 7 label yakni ekspresi senang, sedih, terkejut, netral, marah, jijik dan takut. Model akan diuji dengan dua tahap, yakni testing data dari dataset dan video data yang menunjukkan beberapa ekspresi wajah siswa Teknologi Informasi yang diambil menggunakan platform Zoom Meeting. Hasil akurasi terbaik sebesar 0,97 diperoleh arsitektur VGG16 dengan fungsi optimalisasi Adam dan learning rate 0,0001. Hasil testing data didapatkan nilai f1 score 0,91 dibanding CNN yang hanya mendapatkan nilai di 0,48.
==============================================================================================================================
The COVID19 pandemic has encouraged students and educators to see that technological developments have adapted quickly, one example of which is the application of online learning. In the study conducted, the paradigm shift to online learning has gained relevance in university education and is expected to last even after the pandemic. In this study, the use of Facial Expression Recognition (FER) facial recognition technology will be carried out which is used to recognize or identify students' interest in the material presented based on facial expressions. Therefore, this research will be able to assist educators in knowing student expressions in online learning. In this research, a facial expression classification model will be created using the VGG16 architecture on the FER2013 dataset where there are 7 labels, namely expressions of joy, sadness, surprise, neutral, anger, disgust and fear. The model will be tested in two stages, namely testing data from the dataset and video data showing several facial expressions of Information Technology students taken using the Zoom Meeting platform. The best accuracy results of 0.97 were obtained by the VGG16 architecture with the Adam optimization function and a learning rate of 0.0001. The results of testing the data obtained an f1 score of 0.91 compared to CNN which only got a value of 0.48.

Item Type:	Thesis (Other)
Uncontrolled Keywords:	Deep learning, Ekspresi wajah, Pembelajaran daring, VGG16, Face expression, Online learning
Subjects:	Q Science > Q Science (General) > Q325.5 Machine learning. Support vector machines. T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TA Engineering (General). Civil engineering (General) > TA1650 Face recognition. Optical pattern recognition.
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information Technology > 59201-(S1) Undergraduate Thesis
Depositing User:	Clarissa Fatimah
Date Deposited:	04 Feb 2023 20:55
Last Modified:	06 Feb 2023 01:38
URI:	http://repository.its.ac.id/id/eprint/96211

Actions (login required)

View Item