Skripsi

KLASIFIKASI TEKS BERITA NON-FORMAL DAN FORMAL DALAM BAHASA INDONESIA MENGGUNAKAN ALGORITMA LONG SHORT-TERM MEMORY BERDASARKAN KAMUS BESAR BAHASA INDONESIA

---

This research aims to develop an Indonesian news text classification system to distinguish between formal and non-formal texts. This classification system utilises the Long Short-Term Memory (LSTM) algorithm with reference to the Big Indonesian Dictionary (KBBI). The development process involves text preprocessing techniques, such as data cleaning, lower case, remove punctuation, tokenising, stopword removal and stemming, to improve the quality of the data before it is processed by the LSTM model. The data used consists of 50,000 news texts divided into three parts: 60% training data, 20% validation data, and 20% testing data. The LSTM model was developed with hyperparameters such as RMSProp optimiser, batch size 128, and 40 epochs. The evaluation showed that the LSTM model was able to achieve 81.12% accuracy, 83.65% precision, 92.15% recall, and 87.70% F1-Score. However, the main challenges in this classification include the frequently changing variation of non-formal language, mislabelling in the dataset, limited representative data, risk of overfitting, as well as optimal hyperparameter determination, this system managed to overcome these challenges well.

---

Availability

Detail Information

Series Title

-

Call Number

T1596062024

Publisher

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

Collation

xiii, 57 hlm.; ilus.; 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

621.319 307

Content Type

Text

Media Type

unmediated

Carrier Type

-

Edition

-

Subject(s)

Prodi Teknik Elektro
Bahan Isolasi Listrik

Specific Detail Info

-

Statement of Responsibility

TUTI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment