Penerapan Artificial Neural Network Dalam Mendeteksi Pesan Teks Kurang Pantas Pada Komunikasi Antara Mahasiswa Dan Dosen

Faradhilla, Vanessa Afyta (2024) Penerapan Artificial Neural Network Dalam Mendeteksi Pesan Teks Kurang Pantas Pada Komunikasi Antara Mahasiswa Dan Dosen. Undergraduate thesis, UPN "Veteran" Jawa Timur.

	Text (Cover) 20081010027 - Cover.pdf Download (1MB)
	Text (Bab 1) 20081010027 - Bab 1.pdf Download (126kB)
	Text (Bab 2) 20081010027 - Bab 2.pdf Restricted to Registered users only until 20 September 2026. Download (451kB)
	Text (Bab 3) 20081010027 - Bab 3.pdf Restricted to Registered users only until 20 September 2026. Download (905kB)
	Text (Bab 4) 20081010027 - Bab 4.pdf Restricted to Registered users only until 20 September 2026. Download (1MB)
	Text (Bab 5) 20081010027 - Bab 5.pdf Download (112kB)
	Text (Daftar Pustaka) 20081010027 - Daftar Pustaka.pdf Download (15kB)
	Text (Lampiran) 20081010027 - Lampiran.pdf Restricted to Registered users only until 20 September 2026. Download (25kB)

Abstract

Communication interactions between students and lecturers in an educational environment often occur through text messages, such as messages on the WhatsApp and Telegram platforms. However, the use of increasingly diverse language styles has triggered inappropriate or impolite communication in interacting with lecturers. This is a concern, especially regarding the variety of communication styles used by students. This study aims to implement and evaluate the performance of an Artificial Neural Network model in detecting inappropriate text messages in communication between students and lecturers. The research involves classifying text messages into “appropriate” and “inappropriate” categories, with data taken from real conversations and posts on social media platforms. Preprocessing stages include data cleaning, normalization, and feature extraction using TF-IDF. The ANN model is then trained using Multilayer Perceptron architecture to classify messages based on the labeled data. The research was conducted with 4 test scenarios where the best results in each test will be used in the next test. The results show that the ANN model can effectively identify appropriate and inappropriate messages with an average accuracy of 98% and a loss value of 0.6908. Thus, this ANN model is suitable for use in classifying text messages. Keywords: Student-Faculty Communication, ANN, TF-IDF, Text Message Classification

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
UNSPECIFIED	Mandyartha, Eka Prakarsa	NIDN0725058805	eka_prakarsa.fik@upnjatim.ac.id
UNSPECIFIED	Rizki, Agung Mustika	NIDN0025079302	agung.mustika.fik@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Vanessa Afyta Faradhilla

Date Deposited:

20 Sep 2024 07:35

Last Modified:

20 Sep 2024 07:35

URI:

https://repository.upnjatim.ac.id/id/eprint/29625