Article Per Year (5 Year)
p-Index From 2021 - 2026
0.408
P-Index
This Author published in this journals
All Journal INTECOMS: Journal of Information Technology and Computer Science International Journal Software Engineering and Computer Science (IJSECS)
Makhsun Makhsun
Universitas Pamulang
Computer Science & IT

Published : 2 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 2 Documents
Search

 1 
DAPATKAH MODEL TRANSFORMER MENDETEKSI TOKEN SCAM? SEBUAH STUDI PADA SMART CONTRACT ERC-20 Andhi Saputro; Makhsun Makhsun; Ahmad Musyafa
INTECOMS: Journal of Information Technology and Computer Science Vol. 9 No. 2 (2026): INTECOMS: Journal of Information Technology and Computer Science
Publisher : Institut Penelitian Matematika, Komputer, Keperawatan, Pendidikan dan Ekonomi (IPM2KPE)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31539/h46hgh88

Abstract

Pertumbuhan pesat ekosistem blockchain telah melahirkan ribuan token pada jaringan ERC-20. Fenomena ini mendorong inovasi finansial, namun sekaligus meningkatkan risiko penipuan melalui smart contract yang menyembunyikan mekanisme berbahaya seperti backdoor, blacklist bot, dan manipulasi fee. Penelitian ini mengusulkan pendekatan klasifikasi berbasis Transformer secara end-to-end untuk mendeteksi token scam ERC-20 menggunakan kode sumber Solidity sebagai satu-satunya fitur masukan. Tiga model dievaluasi: CodeBERT (microsoft/codebert-base), RoBERTa (roberta-base), dan GraphCodeBERT (microsoft/graphcodebert-base). Dataset terdiri dari 60.000 kontrak ERC-20 yang diambil dari repositori ASSERT-KTH/DISL, dengan 30.000 kontrak dilabeli secara semi-otomatis menggunakan analisis kode statis berbasis aturan (rule-based) untuk tujuh jenis scam: Honeypot, High Tax, Balance Manipulation, Blacklist, Hidden Owner, Rug Pull, dan Unlimited Mint. Pada klasifikasi biner, GraphCodeBERT mencapai performa terbaik dengan F1-Score 0,9295 dan AUC 0,9808. Pada klasifikasi multilabel, RoBERTa unggul pada F1-Score (0,8681) sementara GraphCodeBERT unggul pada AUC (0,9672). Label Blacklist menjadi tantangan tersendiri dengan F1-Score hanya 0,61–0,64 akibat ketidakseimbangan kelas yang ekstrem. Hasil penelitian membuktikan bahwa representasi kode sumber Solidity melalui model Transformer sudah cukup informatif untuk membedakan kontrak scam dari kontrak legitim secara otomatis.
Nationwide PM2.5 Concentration Prediction in Indonesia Using GRU, GRU-Attention, and BiGRU-Attention Models with Sentinel-5P and ERA5-Land Data Lina Adrianti; Tukiyat Tukiyat; Makhsun Makhsun; Tri Ubaya
International Journal Software Engineering and Computer Science (IJSECS) Vol. 6 No. 2 (2026): AUGUST 2026
Publisher : Lembaga Komunitas Informasi Teknologi Aceh (KITA), Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35870/ijsecs.v6i2.7488

Abstract

Fine particulate matter (PM2.5) pollution has become a serious public health and environmental concern in Indonesia, yet ground-based monitoring stations remain limited and unevenly distributed across the archipelago, restricting comprehensive spatial assessment of pollutant concentrations. This study aimed to develop and comparatively evaluate three deep learning architectures, namely Gated Recurrent Unit (GRU), GRU with Additive Attention (GRU-Attention), and Bidirectional GRU with Additive Attention (BiGRU-Attention), for predicting daily PM2.5 concentrations in Indonesia by integrating Sentinel-5P satellite atmospheric chemistry products and ERA5-Land meteorological reanalysis data. The dataset combined PM2.5 ground-truth measurements from 26 BMKG monitoring stations covering the period 2020–2025, five Sentinel-5P pollutant variables and four ERA5-Land meteorological variables, producing 35,842 cleaned observations and seventeen engineered features. All variables were spatially and temporally aligned, normalized using RobustScaler with log1p target transformation, and reshaped into seven-day sequences using a stratified-station train, validation, and test split with proportions of 70%, 15%, and 15%. The results showed that BiGRU-Attention achieved the best performance with R² of 0.8302, RMSE of 7.4720 µg/m³, MAE of 4.8368 µg/m³, and MAPE of 26.7009%, outperforming GRU-Attention and the baseline GRU. This MAPE is higher than typical single-city PM2.5 models but is consistent with national-scale studies, where low-concentration observations inflate percentage-based errors. The best model was subsequently applied to produce a national daily PM2.5 distribution map, which can help identify regional PM2.5 hotspots, prioritize locations for additional monitoring infrastructure, and inform targeted air quality interventions in regions where ground-based coverage remains sparse.
Co-Authors Ahmad Musyafa Andhi Saputro Lina Adrianti Tri Ubaya Tukiyat Tukiyat