KEYPHRASE EXTRACTION DENGAN MENGGUNAKAN PRE-TRAINED LANGUAGE MODELS BERT DAN TOPIC-GUIDED GRAPH ATTENTION NETWORKS

NABILAH, AINI and Yusliani, Novi and Darmawahyuni, Annisa (2023) KEYPHRASE EXTRACTION DENGAN MENGGUNAKAN PRE-TRAINED LANGUAGE MODELS BERT DAN TOPIC-GUIDED GRAPH ATTENTION NETWORKS. Undergraduate thesis, Sriwijaya University.

	Text RAMA_55201_09021282025067.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (2MB) | Request a copy
	Text RAMA_55201_09021282025067_TURNITIN.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (9MB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_01_front_ref.pdf - Accepted Version Available under License Creative Commons Public Domain Dedication. Download (507kB)
	Text RAMA_55201_09021282025067_0008118205_8968340022_02.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (408kB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_03.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (131kB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_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_09021282025067_0008118205_8968340022_05.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (378kB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_06.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (59kB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_07_ref.pdf - Bibliography Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (148kB) | Request a copy
	Text RAMA_55201_09021282025067_0008118205_8968340022_08_lamp.pdf - Accepted Version Restricted to Repository staff only Available under License Creative Commons Public Domain Dedication. Download (773kB) | Request a copy

Abstract

The increase in the amount of information and documents has made it difficult for individuals to search for information that matches relevant keywords. This poses a challenge in managing and accessing the required information. One solution to address this challenge is the use of Keyphrase Extraction Systems. These systems aim to generate keyphrase that represent the content of documents, enabling users to find documents based on relevant keywords related to the topic they are searching for. This research use an approach to keyphrase extraction that combines the use of pre-trained language models such as BERT and Topic-guided Graph Attention Networks. This method allows for the capture of semantic relationships between words in documents based on their topics. Empirical studies were conducted on a dataset containing 100 accredited scientific journal publications from Sinta 2 and Sinta 3 to evaluate the system's performance. The experimental results show a precision value of 0.058, a recall value of 0.070, and an F1-score of 0.062 for the five generated keywords across the entire tested dataset.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	Keyphrase Extraction, Topic-Guided Graph Attention Networks, Pre-Trained Language Models BERT
Subjects:	T Technology > T Technology (General) > T59.5 Automation
Divisions:	09-Faculty of Computer Science > 55201-Informatics (S1)
Depositing User:	Aini Nabilah
Date Deposited:	05 Jan 2024 04:37
Last Modified:	05 Jan 2024 04:37
URI:	http://repository.unsri.ac.id/id/eprint/137502

Actions (login required)

View Item