Skripsi

KLASIFIKASI EMOSI TEKS SINGKAT MENGGUNAKAN METODE LONG SHORT-TERM MEMORY (LSTM)

---

Penilaian

---

Short text messages, such as those found on social media and messaging platforms, often reflect the emotional expressions of users. Emotion clasification in these texts has significant potential for applications such as sentiment analysis and content personalization. This study aims to develop a short text emotion clasification model using the Long Short-Term Memory (LSTM) method. The dataset used was obtained from the Hugging Face platform, consisting of 18,358 labeled sentences categorized into five emotions: sadness, joy, anger, fear, and surprise. The data underwent a pre-processing phase, followed by word embedding using Word2Vec before being input into the emotion clasification model. The evaluation of the model was carried out under several experimental configurations to determine optimal performance. The best results were achieved with an LSTM configuration using 128 neurons, a dropout of 0.2, a learning rate of 0.001, batch size of 16, and 10 epochs. The developed model achieved an accuracy of 87.61%, a precision of 88.31%, a recall of 87.61%, and an F1-score of 87.81%, indicating its effectiveness in detecting emotions from short text.

---

Availability

Inventory Code	Barcode	Call Number	Location	Status
2507004837	T180651	T1806512025	Central Library (REFERENS)	Available but not for loan - Not for Loan

Detail Information

Series Title

-

Call Number

T1806512025

Publisher

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

Collation

xiii, 73 hlm.; ilus.; tab, 29 cm

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

unmediated

Carrier Type

other (computer)

Edition

-

Subject(s)

Kecerdasan Buatan
Prodi Teknik Informatika

Specific Detail Info

-

Statement of Responsibility

MURZ

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment