Perbandingan Algoritma Decision Tree dan Naive Bayes Classifier pada Analisis Sentimen Twitter Mengenai Kebijakan Penghapusan Kewajiban Skripsi

Idhana, Ilham Ainur (2025) Perbandingan Algoritma Decision Tree dan Naive Bayes Classifier pada Analisis Sentimen Twitter Mengenai Kebijakan Penghapusan Kewajiban Skripsi. Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 19081010099-cover.pdf Download (2MB)
	Text (Bab 1) 19081010099-bab1.pdf Download (87kB)
	Text (Bab 2) 19081010099-bab2.pdf Restricted to Repository staff only until 6 May 2027. Download (321kB)
	Text (Bab 3) 19081010099-bab3.pdf Restricted to Repository staff only until 6 May 2027. Download (429kB)
	Text (Bab 4) 19081010099-bab4.pdf Restricted to Repository staff only until 6 May 2027. Download (5MB)
	Text (Bab 5) 19081010099-bab5.pdf Download (10kB)
	Text (Daftar Pustaka) 19081010099-daftarpustaka.pdf Download (147kB)

Abstract

The policy of removing the thesis obligation as a graduation requirement for S1/D4 students stipulated in Permendikbudristek Number 53 of 2023 has raised various reactions among the public, especially on social media such as Twitter. Therefore, the Decision Tree (DT) Naïve Bayes Classifier (NBC) algorithm is applied to classify public sentiment towards the policy of removing the thesis obligation in Indonesia. The main objective of this research is to evaluate the performance of DT and NBC algorithms in classifying text data into three categories, namely positive, neutral, and negative. The hyperparameter tuning process will be carried out on the DT and NBC models to obtain the best parameters, then use the accuracy and average AUC-ROC score as indicators to determine the most optimal model. The dataset used is thousands of tweets that have gone through preprocessing and labeling stages using a lexicon-based approach and feature extraction using TF-IDF. The results show that negative sentiment is more dominant than positive and neutral sentiment. Model performance evaluation was conducted using accuracy metrics, where the best results were obtained in the Naïve Bayes model with an accuracy of 77.13% and an average AUC-ROC value of 56%, while Decision Tree achieved an accuracy of 74.21% and an average AUC-ROC value of 53%. These results show that Naïve Bayes is superior in the classification of public opinion-based text on social media compared to Decision Tree.

Item Type:

Thesis (Undergraduate)

Contributors:

Contribution	Contributors	NIDN/NIDK	Email
Thesis advisor	Rahmat, Basuki	19690723 2021211 002	basukirahmat.if@upnjatim.ac.id
Thesis advisor	Puspaningrum, Eva Yulia	19890705 2021212 002	evapuspaningrum.if@upnjatim.ac.id

Subjects:

T Technology > T Technology (General)

Divisions:

Faculty of Computer Science > Departemen of Informatics

Depositing User:

Ilham Ainur Idhana

Date Deposited:

06 May 2025 08:35

Last Modified:

06 May 2025 08:35

URI:

https://repository.upnjatim.ac.id/id/eprint/35893