Skripsi

ANALISIS KINERJA MODEL T5 DAN BART DALAM KEYPHRASE GENERATION PADA DATASET ARTIKEL ILMIAH

---

Penilaian

---

Keyphrase Generation is a task in Natural Language Processing that aims to extract keyphrases that represent the main content of a document. The increase in the number of scientific publications has made the manual annotation process inefficient and prone to subjectivity. Therefore, Automated Keyphrase Generation (AKG) has become a relevant approach to support more efficient scientific information management. This study compares the performance of two transformer-based keyphrase generation models, namely Text-to-Text Transfer Transformer (T5) and Bidirectional and Auto-Regressive Transformer (BART). Both models were trained using a fine-tuning approach on 10,000 samples from the KP20K dataset, with 1,000 samples each used for validation and testing. Testing was also conducted on other datasets consisting of other scientific article datasets, namely Inspec, SemEval, NUS, and Krapivin. The performance of both models was evaluated using the F1-Score metric to assess the agreement between the predictions and the reference key phrases. The experimental results show that T5 consistently outperforms BART, with an F1-Score of 21.01% at F1@5 and 22.36% at F1@M.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507003791	T177335	T1773352025	Central Library (Reference)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1773352025

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xiv, 139 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

004.07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment