Skripsi

PEMBANGKITAN PARAFRASA MENGGUNAKAN METODE TRANSFORMER

---

Penilaian

---

Paraphrase generation is a part of text generation or Natural Language Generation (NLG) that aims to create paraphrased sentences with a different words or structure from the input sentence without altering its original meaning. This research builds a paraphrase generation system using the Transformer method. The results of the study indicate that the paraphrase generation model was successfully built using the hyperparameter configuration num_layers = 4, d_model = 128, dff = 512, num_heads = 8, dropout rate = 0.2, with a total of 25 epochs. Based on the BLEU evaluation metric, the model achieved scores of 0.48 on BLEU-1, 0.36 on BLEU-2, 0.28 on BLEU-3, and 0.22 on BLEU-4. Additionally, the model achieves a score of 0.50 on the METEOR evaluation metric and a score of 0.55 on the ROUGE-L evaluation metric. This indicates that the model can provide different word structure variations while maintaining semantic similarity. However, the generated paraphrased sentences sometimes use synonyms that are less contextually accurate in Indonesian due to the dataset being derived from translated English paraphrase sentence pairs.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2407006898	T162006	T1620062024	Central Library (REFERENCE)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1620062024

Publisher

Indralaya : Prodi Teknik Informatika, Fakultas Ilmu Komputer Universitas Sriwijaya., 2024

Collation

xviii, 111 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.350 7

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika
Paraphrasing; Transformer

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment