Skripsi

PENERAPAN NAMED ENTITY RECOGNITION (NER) MENGGUNAKAN ARSITEKTUR LAYOUTLM UNTUK EKSTRAKSI INFORMASI NUTRISI PADA LABEL MAKANAN KEMASAN

---

This study implements the LayoutLM model for the Named Entity Recognition (NER) task to extract nutritional information from packaged food labels. The LayoutLM architecture was chosen for its ability to integrate textual and spatial layout information (bounding boxes), overcoming the limitations of text-only models in processing semi-structured documents. The model was fine-tuned using the openfoodfacts/nutrient-detection-layout dataset across eight scenarios with varying learning rates and batch sizes. Performance was evaluated using precision, recall, and F1-Score, with F1-Score as the primary indicator due to class imbalance. The best result was achieved with the Batch Size 4 and Learning Rate 5e-5 (Scenario 7) configuration, reaching an F1-Score of 92.69%. This confirms LayoutLM's effectiveness for accurate nutritional entity extraction.

---

Availability

Detail Information

Series Title

-

Call Number

T1897262025

Publisher

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

Collation

xvi, 156 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.370 7

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Informasi Nutrisi--Lebel Makanan

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment