Skripsi

KLASIFIKASI BERITA HOAX BAHASA INDONESIA MENGGUNAKAN METODE LSTM (LONG SHORT TERM MEMORY)

---

Penilaian

---

Hoax news poses a serious threat in the current digital environment as it harms society by disseminating false and misleading information. Detecting hoax news is essential to maintain the integrity of information, prevent the spread of misinformation, and safeguard public safety. This research adopts a classification approach to hoax news based on LSTM (Long Short-Term Memory) with the utilization of Word2Vec for word representation. This approach is chosen because LSTM can comprehend the context and sequence of words in a sentence, while Word2Vec provides a meaningful representation of words. By combining the strengths of both methods, this study aims to contribute to improving the detection of Indonesian hoax news. The data used in this study consist of 17,656 records, with 2 classes namely real and hoax. The research findings indicate that the LSTM model with the best configuration, including a learning rate of 0.0001, dropout of 0.2, hidden layer of 64, and batch size of 64, achieves high accuracy in classifying hoax news. The model evaluation using accuracy metric of 94%, recall of 96%, precision of 93%, and f1-score of 94% confirms the reliability of this model in identifying and distinguishing hoax news with a high level of accuracy. Key word : Long Short Term Memory, Word2Vec, Hoax News, News Classification

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2407002618	T143114	T1431142024	Central Library (References)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1431142024

Publisher

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

Collation

xvii, 87 hlm.; ilus.; tab.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

004.07

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

KA

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment