Skripsi

SENTIMENT-BASED KNOWLEDGE DISCOVERY PADA APLIKASI IPUSNAS MENGGUNAKAN METODE MACHINE LEARNING DAN DEEP LEARNING

---

iPusnas is a digital library application with over 1.5 million users, but only received a rating of 2.0. This study conducted a sentiment analysis on 7,596 reviews obtained through web scraping using Google Play Scraper. The data was processed through case folding, data cleaning, tokenization, stopword removal, and stemming, then automatically labeled based on the rating. The dataset was divided in an 80:20 proportion, with 80% for training and 20% for testing. The models tested included Support Vector Machine, Random Forest, CNN, LSTM, and RNN with metrics of accuracy, precision, recall, F1-score, and confusion matrix. CNN and LSTM achieved the highest accuracy of 82%, LSTM obtained the best F1-score of 79%, while RNN recorded the highest recall of 79%. The McNemar test shows significant differences between Random Forest and CNN, Random Forest and CNN, RNN and LSTM, and SVM and Random Forest. Meanwhile, the Random Forest and RNN, CNN and RNN, and CNN and LSTM models show insignificant differences. Keywords : Sentiment Analysis, Deep Learning, iPusnas, Knowledge Discovery, Machine Learning

---

Availability

Detail Information

Series Title

-

Call Number

T1895582025

Publisher

: Prodi Sistem Informasi, Fakultas Ilmu Komputer Universitas Sriwijaya., 2025

Collation

xvii, 100 hlm.; ilus.; tab.; 29 cm.

Language

Indonesia

ISBN/ISSN

-

Classification

006.307

Content Type

Text

Media Type

-

Carrier Type

-

Edition

-

Subject(s)

Prodi Sistem Informasi

Specific Detail Info

-

Statement of Responsibility

MI

Other version/related

File Attachment

Comments

---

You must be logged in to post a comment