Skripsi

CLINICAL NAMED ENTITY RECOGNITION PADA DATA BIOMEDIS MENGGUNAKAN VARIAN MODEL BERT UNTUK KASUS BIOMEDIS

---

Penilaian

---

This study aims to develop a system capable of identifying and classifying medical entities from unstructured biomedical texts, thereby supporting Clinical analysis and health research. The author trained and evaluated three BERT-based models: BioBERT, Clinical BERT, and BlueBERT, for the task of Clinical Named Entity Recognition (CNER). Model performance was measured using precision, recall, and F1-Score metrics on three biomedical datasets: NCBI, BC2GM, and JNLPBA. The results consistently show that BioBERT delivers the best performance across most datasets. On NCBI, BioBERT achieved the highest F1-Score of 93%, outperforming Clinical BERT and BlueBERT (both 91%). In BC2GM, BioBERT also excelled with 91%, while other models reached 90%. The JNLPBA dataset proved more challenging, with BioBERT achieving the highest F1-Score of only 80%. A significant performance improvement, up to 10% in some cases, was observed in the final experiments due to the application of full fine-tuning. This research contributes to identifying the most optimal transformer model for automated information extraction applications in healthcare.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507004329	T179473	T1794732025	Central Library (Reference)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1794732025

Publisher

Indralaya : Prodi Sistem Komputer, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xvii, 117 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

004.07

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Sistem komputer

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment