Search for collections on Repository Universitas Sulawesi Barat

ANALISIS SENTIMEN PENGGUNAAN CHATGPT OPENAI UNTUK PENDIDIKAN DALAM PANDANGAN MAHASISWA UNSULBAR MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM) SENTIMENT ANALYSIS ON THE USE OF OPENAI'S CHATGPT FOR EDUCATIONAL PURPOSES FROM THE PERSPECTIVE OF UNSULBAR STUDENTS USING THE SUPPORT VECTOR MACHINE (SVM) METHOD

ALFIANA, SULMA (2025) ANALISIS SENTIMEN PENGGUNAAN CHATGPT OPENAI UNTUK PENDIDIKAN DALAM PANDANGAN MAHASISWA UNSULBAR MENGGUNAKAN METODE SUPPORT VECTOR MACHINE (SVM) SENTIMENT ANALYSIS ON THE USE OF OPENAI'S CHATGPT FOR EDUCATIONAL PURPOSES FROM THE PERSPECTIVE OF UNSULBAR STUDENTS USING THE SUPPORT VECTOR MACHINE (SVM) METHOD. Diploma thesis, UNIVERSITAS SULAWESI BARAT.

[thumbnail of ALFIANA SULMA.pdf] Text
ALFIANA SULMA.pdf
Restricted to Repository staff only

Download (3MB)
[thumbnail of ALFIANA SULMA (pdf.io).pdf] Text
ALFIANA SULMA (pdf.io).pdf

Download (2MB)

Abstract

Kecerdasan buatan seperti ChatGPT darı OpenAl memberikan dampak signifikan dalam dunia pendidikan, khususnya mahasiswa. ChatGPT dapat
membantu mahasiswa dalam menyelesaikan tugas, mencari referensi, dan memahami materi, namun juga menimbulkan kekhawatiran terkait plagiarisme dan menurunnya kreativitas belajar. Penelitian ini bertujuan untuk menganalisis sentimen mahasiswa Fakultas Teknik Universitas Sulawesi Barat terhadap penggunaan ChatGPT dalam pendidikan, serta mengevaluası performa klasifikasi menggunakan algoritma Support Vector Machine (SVM). Metode yang digunakan pengumpulan data melalui kuesioner terbuka, preprocessing teks (case folding, tokenisasi, normalisasi, filtering, stemming), pelabelan data menggunakan kamus lexicon (positif, negatif, netral), pembobotan kata dengan TF-IDF serta klasifikasi dengan SVM. Dataset berjumlah 501 tanggapan mahasiswa dengan distribusi sentimen negatif (382), positif (69), dan netral (50). Data dibagi menggunakan rasio 80:20 untuk pelatihan dan pengujian. Hasil pengujian menunjukkan bahwa model SVM menghasilkan akurasi sebesar 83.17%. Kelas sentimen negatif memiliki fl-score tertinggi sebesar 0,91, diikuti oleh kelas positif dengan f1-score 0,67. dan netral dengan fl-score 0,14. Ketidakseimbangan jumlah data antar kelas memengaruhi performa model, di mana kelas dengan data terbanyak (negatif) dikenali lebih baik dibandingkan kelas lainnya. Penelitian ini menunjukkan bahwa mayoritas mahasiswa memiliki pandangan negatif terhadap penggunaan ChatGPT dalam pendidikan.
Artificial intelligence such as ChatGPT from OpenAI has a significant impact in the field of education, particularly among university students. ChatGPT can assist students in completing assignments, finding references, and understanding materials; however, it also raises concerns related to plagiarism and decreased learning creativity. This study aims to analyze the sentiments of students from the Faculty of Engineering, Universitas Sulawesi Barat, regarding the use of ChatGPT in education, as well as to evaluate the performance of classification using the Support Vector Machine (SVM) algorithm.The method used includes data collection through open-ended questionnaires, text preprocessing (case folding, tokenization, normalization, filtering, stemming), data labeling using a lexicon dictionary (positive, negative, neutral), word weighting with TF-IDF, and classification with SVM. The dataset consists of 501 student responses with sentiment distribution as follows: negative (382), positive (69), and neutral (50). The data were split using an 80:20 ratio for training and testing.The testing results show that the SVM model achieved an accuracy of 83.17%. The negative sentiment class
obtained the highest f1-score of 0.91, followed by the positive class with an f1- score of 0.67, and the neutral class with an f1-score of 0.14. The imbalance in data distribution among classes affected the model’s performance, where the class with the largest number of data (negative) was recognized more effectively than the others. This study indicates that the majority of students hold a negative
perspective toward the use of ChatGPT in education.

Item Type: Thesis (Diploma)
Uncontrolled Keywords: ChatGPT, Analisis Sentimen, Support Vector Machine. Pendidikan. TF-IDF. ChatGPT, Sentiment Analysis. Support Vector Machine, Education, TF-IDF.
Subjects: FAKULTAS TEKNIK > Informatika
Divisions: Fakultas Teknik
Depositing User: Unnamed user with email aryatiunsulbar@gmail.com
Date Deposited: 13 Oct 2025 06:29
Last Modified: 13 Oct 2025 06:29
URI: https://repository.unsulbar.ac.id/id/eprint/2374

Actions (login required)

View Item
View Item