Article Per Year (5 Year)
p-Index From 2021 - 2026
0.817
P-Index
This Author published in this journals
All Journal JSM (Jurnal SIFO Mikroskil) CommIT (Communication & Information Technology) Journal of Computer Science and Informatics Engineering
Erlina Halim
Universitas Mikroskil
Computer Science & IT Decision Sciences, Operations Research & Management

Published : 4 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 4 Documents
Search

 1 
Sentimen Analisis Opini Pembeli pada Aplikasi Shopee Berbasis N-Gram Lexicon Erlina Halim; Arwin Halim; Andriana Sunjaya; Novresia Sunjaya
Jurnal SIFO Mikroskil Vol 23, No 2 (2022): JSM VOLUME 23 NOMOR 2 TAHUN 2022
Publisher : Fakultas Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55601/jsm.v23i2.896

Abstract

Perubahan perilaku konsumen dalam melakukan pembelian di e-commerce, telah mengubah cara konsumen dalam memutuskan membeli barang. Konsumen akan menilai berdasarkan gambar, rating dan opini dari pembeli sebelumnya. Namun konsumen harus membaca semua opini untuk mendapatkan hasil yang akurat. Solusi yang ditawarkan adalah dengan menyediakan fitur pengolahan opini. Namun opini yang ditulis tidak terstruktur, terdapat kata yang tidak lengkap, emoticon, gambar, serta menggunakan kata tidak baku menjadi tantangan dalam mengolah opini. Corpus yang digunakan dalam penelitian ini diperoleh dari penelitian sebelumnya yang berjumlah 124.606 dengan atribut yang tersedia yaitu itemname, date, starrate dan opini. Penambahan corpus dilakukan dengan membangun program untuk menarik data opini sebanyak 95.731 sehingga total opini yang diproses sebanyak 220.337. Proses ekstraksi opini menggunakan pendekatan lexicon dengan peningkatan penambahan proses opinion representation dan n-gram lexicon untuk menentukan arah sentimen. Peningkatan proses pre-processing dengan penerapan opinion representation berbasis Bahasa Indonesia pada kasus review e-commerce. Peningkatan klasifikasi dengan memperhatikan semantik kata dengan menerapkan n-gram. Opini hasil klasifikasi menghasilkan kelas positif, netral, dan negatif. Klasifikasi opini akan dibandingkan dengan klasifikasi rating untuk mengetahui tingkat keakuratan. Hasil dari evaluasi diperoleh akurasi sebesar 78%. Terjadi peningkatan akurasi sebesar 2% dibanding dengan ekstraksi opini tanpa adanya proses opinion representation dan n-gram.
Analisis Dan Perancangan Aplikasi Pengelolaan Dan Pemanfaatan Bank Sampah Di Kota Medan Gidion Sagala; Desi Rahayu; Nadila Afari; Erlina Halim; Arwin Halim
Jurnal SIFO Mikroskil Vol 24, No 1 (2023): JSM VOLUME 24 NOMOR 1 TAHUN 2023
Publisher : Fakultas Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55601/jsm.v24i1.966

Abstract

Salah satu masalah sosial terkait lingkungan yang sering ditemui di masyarakat adalah banyaknya timbunan sampah, salah satunya di perkotaan di Indonesia. Hal tersebut ditandai dengan rata-rata sampah yang dibuang sebanyak 0,85 kg setiap hari. Hanya 41% sudah diangkut dan diolah dengan baik. Sisanya sekitar 300.000 ton sampah tidak diangkut dan mencemari lingkungan. Dari permasalahan tersebut maka dirancanglah suatu aplikasi bank sampah berbasis mobile yang dapat meningkatkan kegiatan daur ulang serta mengurangi volume sampah agar mempermudah pengelolaan dan pemanfaatan sampah. Aplikasi ini menyediakan wadah tempat penjualan dan promosi untuk mengelola dan memanfaatkan sampah menjadi barang yang bernilai. Di aplikasi ini juga terdapat wadah informasi tentang seputar pengetahuan cara pengelolaan dan pemanfaatan sampah. Kemudian terdapat wadah untuk membantu para masyarakat yang membutuhkan dengan cara berdonasi dan mendapatkan point/reward dari bermain game yang dapat ditukarkan menjadi uang.
Fine-Tuning Hybrid Deep Learning for Sentiment Analysis of Indonesian Product Reviews Arwin Halim; Roni Yunis; Erlina Halim
CommIT (Communication and Information Technology) Journal Vol. 20 No. 1 (2026): CommIT Journal (in press)
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar

Abstract

The research aims to build a hybrid deep learning model for sentiment analysis of Indonesian ecommerce product reviews, which represent the expressed opinions of customers. A major challenge in the domain is the presence of non-standard language and highly imbalanced sentiment classes, which hinder accurate classification. Most existing Indonesian sentiment analysis studies rely on relatively small and balanced datasets and primarily use attention mechanisms, an ensemble model, as well as a sequential fusion method. In the research, a large-scale dataset of Indonesian product reviews is collected from the largest e-commerce site in the country. The dataset consists of review text and corresponding product ratings. After preprocessing, semantic features are extracted using a pre-trained Indonesia Bidirectional Encoder Representations from Transformers (IndoBERT) model. The features are then fed into a hybrid model combining Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) layers through parallel feature-level fusion. Model hyperparameters are optimized using the Tree-Structured Parzen Estimator (TPE), while data imbalance is addressed through resampling methods. Regularization strategies are also applied to mitigate overfitting, and the model is evaluated using stratified k-fold cross-validation. The model hyperparameters are validated using a learning curve, showing a stable and consistent curve following the trend. The results show that the hybrid CNN-LSTM model, combined with Support Vector Machine Synthetic Minority Oversampling Technique (SVMSMOTE), achieves superior performance in distinguishing positive and negative reviews. This outcome reaches Receiver Operating Characteristic - Area Under the Curve (ROC AUC) score of 92.48%, outperforming baseline and conventional machine learning models. These results also show good generalization ability, characterized by consistent values with a very low standard deviation of 0.0009 for each fold.
Development of an Integrated Web-Based School Information System at SMK Negeri 1 Lumban Julu Dhea Romantika Marpaung; Muhammad Syahputra; Rizky Fadillah; Sunario Megawan; Erlina Halim
Journal of Computer Science and Informatics Engineering Vol 5 No 3 (2026): July
Publisher : Ali Institute of Research and Publication

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55537/cosie.v5i3.1752

Abstract

Digital technology encourages schools to provide faster, more transparent, and more structured information and academic administration services. SMK Negeri 1 Lumban Julu did not yet have an integrated information system combining a public website and a role-based academic portal. This study aims to develop a web-based school information system to support school information delivery and academic data management. The system was developed using the Scrum method through requirements analysis, product backlog preparation, design, sprint implementation, and testing. The developed system provides a public website for school information, publications, departments, extracurricular activities, and tracer study, as well as an academic portal for staff, teachers, and students. Testing was conducted using blackbox testing by three prospective users, consisting of one staff member, one teacher, and one student. Testing covered 260 scenarios, consisting of 198 staff scenarios, 41 teacher scenarios, and 21 student scenarios. All scenarios were successfully executed, resulting in a functional success rate of 100%. These results indicate that the system can support integrated school information and academic data management.
Co-Authors Andriana Sunjaya Arwin Halim Desi Rahayu Dhea Romantika Marpaung Gidion Sagala Muhammad Syahputra Nadila Afari Novresia Sunjaya Rizky Fadillah Roni Yunis Sunario Megawan