PERINGKASAN ABSTRAKTIF PADA DIALOG MENGGUNAKAN FINE-TUNING PRE-TRAINED LANGUAGE MODEL BART

RAMADHANI, HANIF SYAHRI and Rodiah, Desty (2025) PERINGKASAN ABSTRAKTIF PADA DIALOG MENGGUNAKAN FINE-TUNING PRE-TRAINED LANGUAGE MODEL BART. Undergraduate thesis, Sriwijaya University.

	Image RAMA_55201_09021282126042_cover.jpg - Cover Image Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (204kB) | Request a copy
	Text RAMA_55201_09021282126042.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (6MB) | Request a copy
	Text RAMA_55201_09021282126042_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (1MB)
	Text RAMA_55201_09021282126042_0021128905_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (515kB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (315kB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_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_09021282126042_0021128905_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (212kB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (187kB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (191kB) | Request a copy
	Text RAMA_55201_09021282126042_0021128905_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (4MB) | Request a copy

Abstract

The massive increase in communication over digital platforms generates conversation data that contains a variety of important information and requires effective analysis methods to filter the information, one of which is abstractive summarization. This technique uses a natural language processing (NLP) approach to reorganize dialogue information into a more concise and coherent form. This research aims to develop an abstractive dialogue summarization system using pre-trained BART (Bidirectional and Auto-Regressive Transformers) models that are fine-tuned with a sequence-to-sequence approach on the DialogSum dataset. The dataset consists of 14,460 English dialogues, split for training and testing with 7:3 ratio. Evaluation using the ROUGE metrics shows that the average values on ROUGE-1 are precision 53.16%, recall 49.90% and f1-score 50.44%. On ROUGE-2, precision was 25.66%, recall 24.95% and f1-score 24.76%. The results show that the system is able to generate dialogue summaries with the highest performance on the ROUGE-1 metric which measures the number of individual words absorbed, in line with the focus of abstractive summarization which performs new sentence formation through paraphrasing.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Peringkas Abstraktif, Dialogue Summarization, BART, Natural Language Processing, sequence-to-sequence, ROUGE Score
Subjects:	P Language and Literature > P Philology. Linguistics > P98-98.5 Computational linguistics. Natural language processing Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation. T Technology > T Technology (General) > T1-995 Technology (General)
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Hanif Syahri Ramadhani
Date Deposited:	19 Feb 2025 06:23
Last Modified:	19 Feb 2025 06:23
URI:	http://repository.unsri.ac.id/id/eprint/167650

Actions (login required)

View Item