Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal JiTEKH (Jurnal Ilmiah Teknologi Harapan)
Ndhogal Rakasiwi Dhofisa
Program Studi Pendidikan Teknologi Informasi, STKIP PGRI Situbondo, Situbondo
Civil Engineering, Building, Construction & Architecture Computer Science & IT Electrical & Electronics Engineering Engineering Mechanical Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Sistem Analisis Sentimen Ulasan Produk Tokopedia Menggunakan Metode VGG16 Untuk Prediksi Rekomendasi Produk Sesuai Minat Pengguna Ndhogal Rakasiwi Dhofisa; Nur Azizah; Rahmat Shofan Razaqi
JiTEKH Vol. 14 No. 2 (2026): September 2026
Publisher : Universitas Harapan Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35447/jitekh.v14i2.1375

Abstract

Tokopedia, as an e-commerce platform, provides a large number of product reviews containing user opinions on product quality, seller service, and delivery processes. These reviews can be used to assess customer satisfaction and serve as important signals in developing product recommendations. This study proposes a Tokopedia product review sentiment analysis system using a deep learning approach based on Convolutional Neural Networks (CNN) using the VGG16 architecture. To adapt the image-oriented VGG16 to text data, text reviews are rendered into 224x224 images (text-to-image), then VGG16 transfer learning is performed to classify sentiment into positive, negative, and neutral. The resulting sentiment scores are then combined with user interest profiles (frequently viewed and purchased categories) to predict more relevant product recommendations. The system stages include data collection, sentiment labeling, preprocessing, review image formation, VGG16 model training, and evaluation using a confusion matrix and metrics such as accuracy, precision, recall, and F1-score. The system implementation is expected to help users find products that match their interests with the support of more objective review sentiment.
Co-Authors Nur Azizah Rahmat Shofan Razaqi