KLASIFIKASI EMOSI PADA TEKS BERBAHASA INDONESIA DENGAN FINE-TUNING INDOBERT

SYAMSIDI, IMAN CARRAZI and Abdiansah, Abdiansah (2025) KLASIFIKASI EMOSI PADA TEKS BERBAHASA INDONESIA DENGAN FINE-TUNING INDOBERT. Undergraduate thesis, Sriwijaya University.

Preview	Image RAMA_55201_09021182126009_cover.jpg - Cover Image Available under License Creative Commons Public Domain Dedication. Download (226kB) | Preview
	Text RAMA_55201_09021182126009.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (3MB) | Request a copy
	Text RAMA_55201_09021182126009_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (8MB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021182126009_0001108401_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (629kB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (438kB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_04.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (1MB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (210kB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (229kB) | Request a copy
	Text RAMA_55201_09021182126009_0001108401_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (118kB) | Request a copy

Abstract

The increasing use of social media has led to significant growth in Indonesian text data, creating complexity in emotion identification and classification tasks. To address this challenge, this study develops an emotion classification system using the fine-tuning method on the IndoBERT model. This research aims to classify emotions in Indonesian text using IndoBERT fine-tuning. The study utilizes two types of datasets: an imbalanced dataset comprising 4,403 tweets sourced from IndoNLU and a balanced dataset containing 5,600 tweets combined from IndoNLU and Riccosan et al. (2022). The fine-tuning process is divided into eight scenarios, combining different dataset types, batch sizes, and learning rates. The results demonstrate that the fine-tuned IndoBERT model achieves optimal performance on the balanced dataset with 78.55% accuracy, 78,55% recall, 78,64% precision and 78.46% f1-score using a learning rate of 4e-6 and batch size of 32. For the imbalanced dataset, the model attains 75% accuracy, 75,26% recall, 75,60% precision and 75.46% f1-score with a learning rate of 4e-6 and batch size of 16.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	media sosial, klasifikasi emosi, IndoBERT, fine-tuning, accuracy, f1-score
Subjects:	Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Iman Carrazi Syamsidi
Date Deposited:	24 Mar 2025 04:23
Last Modified:	24 Mar 2025 04:23
URI:	http://repository.unsri.ac.id/id/eprint/169992

Actions (login required)

View Item