cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Mustakim
Contact Email
officialmalcom.irpi@gmail.com
Phone
+6285275359942
Journal Mail Official
malcom@irpi.or.id
Editorial Address
INSTITUT RISET DAN PUBLIKASI INDONESIA Jl. Tuah Karya Ujung C7. Kel. Tuah Madani Kec. Tampan Kota Pekanbaru - Riau
Location
Kota pekanbaru,
Riau
INDONESIA
Malcom: Indonesian Journal of Machine Learning and Computer Science
Published by Institut Riset dan Publikasi Indonesia
ISSN : 27972313     EISSN : 27758575     DOI : -
Core Subject : Science,
Computer Science & IT
MALCOM: Indonesian Journal of Machine Learning and Computer Science is a scientific journal published by the Institut Riset dan Publikasi Indonesia (IRPI) in collaboration with several Universities throughout Riau and Indonesia. MALCOM will be published 2 (two) times a year, April and October, each edition containing 10 (Ten) articles. Articles may be written in Indonesian or English. articles are original research results with a maximum plagiarism of 15%. Articles submitted to MALCOM will be reviewed by at least 2 (two) reviewers. The submitted article must meet the assessment criteria and in accordance with the instructions and templates provided by MALCOM. The author should upload the Statement of Intellectual/ Copyright Rights when submitting the manuscript. Papers must be submitted via the Open Journal System (OJS) in .doc or .docx format. The entire process until MALCOM is published will be free of charge. MALCOM is registered in National Library with Number International Standard Serial Number (ISSN) Printed: 2797-2313 and Online 2775-8575. Focus and scope of MALCOM includes Data Mining, Data Science, Artificial Intelligence, Computational Intelligence, Natural Language Processing, Big Data Analytic, Computer Vision, Expert System, Text and Web Mining, Parallel Processing, Intelligence System, Decision Support System and Software Engineering
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 581 Documents
 54   55   56   57   58 
IndoBERT vs SVM: Analisis Sentimen Komentar Penonton Film Penyalin Cahaya: IndoBERT vs SVM: Sentiment Analysis of Audience Comments on the Film Photocopier Rahmadi, Daffa Junichi; Jaya, Safitri
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2784

Abstract

Penelitian ini membuktikan keunggulan Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT) dibandingkan dengan Support Vector Machine (SVM) dalam analisis sentimen pada data komentar di platform digital, khususnya terkait isu yang diangkat dalam film Penyalin Cahaya. Data untuk penelitian ini diperoleh melalui scraping dari lima platform digital, dengan total 12.473 komentar. Tahap preprocessing yang diterapkan meliputi cleaning, normalization, tokenization, penghapusan stopword, dan stemming. Untuk mengatasi ketidakseimbangan data, penelitian ini menggunakan metode Random Over-Sampling (ROS) dan Synthetic Minority Over-Sampling Technique (SMOTE). Hasil evaluasi menunjukkan bahwa IndoBERT memperoleh nilai F1-Score 0.80, sedangkan SVM memperoleh 0.74. Meskipun kedua model memiliki nilai AUC sebesar 0.94. IndoBERT memiliki performa yang lebih baik dalam memahami konteks kalimat, khususnya pada sentimen netral dan negatif. IndoBERT juga lebih unggul dalam menghasilkan prediksi yang lebih akurat dan seimbang pada setiap kategori sentimen. Hasil analisis juga memperlihatkan bahwa opini publik didominasi oleh sentimen positif yang mencerminkan apresiasi terhadap film serta dukungan terhadap isu yang diangkat.
Sistem Pendukung Keputusan Evaluasi Kinerja Dosen Metode SAW Berbasis Data Akademik dan Umpan Balik Mahasiswa: Decision Support System for Lecturer Performance Evaluation Using the SAW Method Based on Academic Data and Student Feedback Dzakiyyah, Syifa Ghina; Yani, Putri; Hikmah, Maulida; Triyono, Gandung
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2787

Abstract

Evaluasi kinerja dosen di institusi pendidikan tinggi kerap terkendala oleh penilaian yang subjektif dan multikriteria. Penelitian ini bertujuan merancang Sistem Pendukung Keputusan (SPK) menggunakan metode Simple Additive Weighting (SAW) untuk mengagregasi parameter yang heterogen secara objektif. Sistem ini mengintegrasikan data akademik internal (kehadiran dan publikasi ilmiah) dengan data umpan balik dari 286 mahasiswa (kualitas pengajaran dan penguasaan materi). Bobot setiap kriteria ditentukan melalui pendekatan Rank Order Centroid (ROC). Hasil pengujian terhadap 50 alternatif dosen menunjukkan sistem berhasil melakukan pemeringkatan dengan presisi, di mana aspek kualitas pengajaran dan produktivitas riset teridentifikasi sebagai faktor penentu utama (performance drivers). Keandalan model ini kemudian divalidasi dengan membandingkan luaran sistem terhadap expert judgment (penilaian holistik Ketua Program Studi) menggunakan 20 sampel uji. Evaluasi tersebut membuktikan sistem memiliki tingkat kesesuaian (Agreement Rate) sebesar 95% dalam mengidentifikasi dosen berkinerja tinggi, sehingga sangat layak diimplementasikan sebagai dasar pengambilan keputusan strategis bagi manajemen.
Klasifikasi Opini Publik Berbasis Aspek terhadap Affiliate Marketing di TikTok Menggunakan Fine-Tuned IndoBERT: Classification of Public Opinion on Affiliate Marketing on TikTok Based on Specific Aspects Using Fine-Tuned IndoBERT Oktafiyana, Vina; Yudhistira, Aditia
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2804

Abstract

Affiliate marketing pada TikTok berkembang pesat sebagai strategi pemasaran digital berbasis komisi yang melibatkan kreator konten dalam mempromosikan produk kepada pengguna. Aktivitas ini memunculkan berbagai opini publik terkait kualitas produk, kredibilitas kreator, harga dan promosi, pengalaman belanja, serta kualitas konten yang dibagikan melalui media sosial. Penelitian ini bertujuan menganalisis sentimen pengguna terhadap praktik affiliate marketing menggunakan pendekatan aspect-based sentiment classification (ABSC) berbasis Indonesian Bidirectional Encoder Representations from Transformers (IndoBERT). Sebanyak 10.000 komentar TikTok berbahasa Indonesia dikumpulkan dan diproses melalui tahapan preprocessing, deduplikasi, identifikasi aspek menggunakan keyword matching dan Term Frequency — Inverse Document Frequency (TF-IDF) cosine similarity, serta pelabelan sentimen menggunakan pretrained IndoBERT. Kebaruan penelitian ini terletak pada penerapan pendekatan terintegrasi antara identifikasi aspek berbasis keyword matching dan TF-IDF cosine similarity dengan model IndoBERT yang di-fine-tune untuk klasifikasi sentimen berbasis aspek pada komentar TikTok berbahasa Indonesia terkait affiliate marketing. Pendekatan ini mampu memberikan analisis opini yang lebih kontekstual pada domain affiliate marketing yang masih jarang dibahas dalam penelitian sebelumnya. Model kemudian di-fine-tune menggunakan weighted cross-entropy loss untuk menangani ketidakseimbangan kelas pada data. Hasil pengujian menunjukkan akurasi sebesar 89,79%, weighted F1-score sebesar 89,74%, dan macro F1-score sebesar 82,29%. Evaluasi menggunakan stratified 5-fold cross-validation menghasilkan rata-rata akurasi sebesar 91,03%  
Pengembangan Sistem E-learning Responsif untuk Sekolah Menengah Kejuruan Berbasis Laravel dengan SCRUM: Responsive E-learning System Development for Vocational High School Using Laravel and SCRUM Hasibuan, Rajimar Suhal; Amir, Faisal; Oriyasmi, Fadhilah; Fadilillah, Fadli; Mardoni, Mardoni
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2828

Abstract

Penelitian ini dilatarbelakangi oleh masih terbatasnya sistem pembelajaran daring yang terstruktur di SMK Hasanah Pekanbaru. Selama ini guru hanya mengandalkan Whatsapp untuk berbagi materi dan tugas, sehingga tidak ada pelacakan nilai, pengumpulan tugas terpusat, maupun forum diskusi akademik. Tujuan penelitian adalah membangun sistem E-learning yang responsif terhadap perubahan kebutuhan pengguna dengan menerapkan metode Agile Scrum dan framework Laravel. Proses pengembangan dilakukan dalam 4 sprint (masing-masing 2 minggu), melibatkan 5 guru dan 10 siswa sebagai product owner dalam setiap sprint review. Fitur yang dihasilkan meliputi manajemen materi, pengumpulan tugas dengan batas waktu, forum diskusi per mata pelajaran, ujian Online, serta laporan nilai otomatis. Pengujian black-box pada 27 skenario (termasuk 10 skenario negatif) menghasilkan tingkat keberhasilan 96,3% setelah perbaikan pada kasus upload file kosong. User Acceptance Test (UAT) dengan 40 responden (15 guru dan 25 siswa) menghasilkan tingkat kepuasan rata-rata 90,8%, dengan rincian aspek pengurangan keterlambatan tugas 94%, kemudahan proses belajar mengajar 93%, kepuasan keseluruhan 92%, kemudahan antarmuka 89%, dan efektivitas forum diskusi 86%. Penelitian ini berkontribusi pada penerapan Scrum yang terdokumentasi secara iteratif di lingkungan SMK, serta menyediakan fitur pengingat deadline dan rekap nilai otomatis yang belum ada pada Learning Management System (LMS) umum. Sistem dinyatakan layak digunakan secara riil.
A Segmentation-Aware Isolation Forest Framework for Fraud Investigation Prioritization in QRIS Transactions Kamal, Ahmad Rijaluddin Nisbil; Lukitosari, Valeriana
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2832

Abstract

Fraud detection in Quick Response Code Indonesian Standard (QRIS) transaction environments is challenging because fraud labels are often limited, delayed, or incomplete, making it difficult to apply supervised learning effectively. In addition, high transaction volumes create operational constraints for investigation teams, requiring an efficient mechanism to prioritize high-risk merchants for review. To address this problem, this study proposes a segmentation-aware Isolation Forest framework for fraud investigation prioritization in QRIS transactions. Transaction-level data from a national bank in Indonesia were transformed into merchant-level behavioral representations using 38 engineered features capturing transaction intensity, temporal patterns, nominal concentration, burst behavior, and transactional irregularities. Anomaly scores were then used as ranking signals for prioritizing selective investigation. Model performance was evaluated using proxy-based ranking metrics, capacity simulation investigations, selective alert optimization, transaction-level interpretability, and expert validation. The results show that only 1.03% of merchants were classified as very high anomalies, while the model achieved strong ranking performance with an Area Under the Curve (AUC) of 0.8794, Precision@Top 5% 0.3306, and Lift of 6.5942. Selective alerting reduced investigation workload by up to 80.24%, with higher precision and lower FAR, although at the cost of reduced recall. Expert validation further indicated 94.12% prioritized merchants were operationally relevant.
Hybrid Learning for Automated E-commerce Churn Prediction with XGBoost and SHAP Henry, Amir Acalapati; Arribathi, Abdul Hamid; Henderi, Henderi
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2833

Abstract

Customer retention is a critical challenge in the e-commerce industry, yet many platforms frequently suffer from a lack of explicit labels to identify potential defectors (churn). This research proposes a hybrid unsupervised-supervised learning framework to perform automated churn classification using the high-dimensional Olist Brazilian E-commerce dataset. The first stage employs K-Means Clustering to objectively generate churn labels based on Recency, Frequency, and Monetary (RFM) features. The second stage applies an Extreme Gradient Boosting (XGBoost) model, augmented by the Synthetic Minority Over-sampling Technique (SMOTE), to address the inherent class imbalance typical of transactional data. Experimental results demonstrate that the proposed model achieved a robust accuracy of 72% with a churn recall rate of 69%. Interpretability analysis using SHapley Additive exPlanations (SHAP) revealed that delivery duration and customer review scores are the most dominant predictors of churn, significantly outweighing financial metrics. These findings contribute a novel integration of automated labeling and model transparency, enabling e-commerce managers to implement proactive, data-driven customer retention strategies.
Sentiment Analysis of Cyberbullying Against Children on Social Media: A Naïve Bayes Approach Fatin, Alfindian Kurnia; Adhitama, Rifki; Istighosah, Maie
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2858

Abstract

Cyberbullying involving children has become a major concern on social media, generating diverse public responses. This study aims to analyze public sentiment toward child cyberbullying cases on social media and to evaluate the effectiveness of data-balancing techniques in improving the performance of a Naïve Bayes (NB) classifier. The motivation for using NB lies in its computational efficiency, strong performance in text classification, and suitability for high-dimensional Term Frequency–Inverse Document Frequency (TF-IDF) features. A dataset consisting of 3,137 comments was collected from X/Twitter and processed through text preparation, lexicon-based sentiment labeling using Indonesian Sentiment Lexicon (InSet), and TF-IDF feature representation. Three balancing scenarios, namely Random Oversampling, Random Undersampling, and Non-Balancing, were compared by varying the proportion of training and testing samples across four split ratios: 90:10, 80:20, 70:30, and 60:40. The experimental results show that balancing techniques contribute to classification effectiveness, with Random Oversampling producing the best outcome. The highest accuracy of 86.36% was achieved with Random Oversampling and a 90:10 training-test split, outperforming Random Undersampling and Non-Balancing. Furthermore, the sentiment distribution indicates that negative sentiment dominates public responses to child cyberbullying incidents. 
Prediksi Pengangguran di Nusa Tenggara Barat Menggunakan Regresi Linier Berganda Berbasis Supervised Learning: Predicting Unemployment in West Nusa Tenggara Using Supervised Machine Learning-Based Multiple Linear Regression Rukmiati, Rukmiati; Mutmainah, Siti; Khairunnas, Khairunnas
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2860

Abstract

Tingkat Pengangguran Terbuka (TPT) merupakan salah satu indikator penting dalam mengukur kondisi ketenagakerjaan dan kesejahteraan masyarakat. Penelitian ini bertujuan untuk menganalisis pengaruh Indeks Pembangunan Manusia (IPM), Tingkat Kemiskinan, dan Inflasi terhadap TPT di Provinsi Nusa Tenggara Barat serta membangun model prediksi menggunakan Regresi Linear Berganda. Data yang digunakan merupakan data sekunder yang diperoleh dari Badan Pusat Statistik (BPS) Provinsi Nusa Tenggara Barat. Tahapan penelitian meliputi pengumpulan data, preprocessing, analisis data eksploratif (EDA), pengujian asumsi klasik, pemodelan, dan evaluasi model. Hasil pengujian asumsi klasik menunjukkan bahwa model telah memenuhi asumsi normalitas, multikolinearitas, heteroskedastisitas, dan autokorelasi. Untuk memperkuat kontribusi machine learning, dilakukan perbandingan antara Regresi Linear Berganda dan Decision Tree Regression. Hasil evaluasi menunjukkan bahwa Regresi Linear Berganda memperoleh nilai MSE sebesar 0,4357, RMSE sebesar 0,6601, dan R² sebesar 0,3226, sedangkan Decision Tree Regression memperoleh nilai MSE sebesar 0,5442, RMSE sebesar 0,7377, dan R² sebesar 0,1540. Berdasarkan hasil tersebut, Regresi Linear Berganda menghasilkan tingkat kesalahan yang lebih rendah dan kemampuan prediksi yang lebih baik dibandingkan dengan Decision Tree Regression. Dengan demikian, Regresi Linear Berganda dipilih sebagai model terbaik untuk memprediksi Tingkat Pengangguran Terbuka di Provinsi Nusa Tenggara Barat.
Cluster-Based Random Forest Regression for Internet of Things Bandwidth Prediction Using Leakage-Free Features Hutagalung, Naek Parulian; Tutuko, Bambang; Zarkasi, Ahmad
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2894

Abstract

Heterogeneous Internet of Things (IoT) deployments generate highly variable network traffic, yet most existing bandwidth management approaches rely on static provisioning or device-agnostic models that fail to capture per-device traffic heterogeneity. This study proposes a hybrid machine learning framework combining K-Means clustering and Random Forest regression to enable cluster-aware, leakage-free bandwidth consumption prediction for IoT environments. Using 994,145 flow records from the UNSW HomeNet dataset, the Log1p transformation was applied to address heavy-tailed traffic distributions, and direct bandwidth-derived features were strictly excluded to prevent data leakage. Devices were segmented into three traffic profiles (Low, Medium, and High), validated by a Silhouette Score of 0.4153 and a Davies–Bouldin Index of 1.0357. Within the structural constraints of this single-dataset, controlled setting, Random Forest regression achieved R² values above 0.99 across all clusters, with stable cross-validated performance, substantially outperforming Linear Regression, which achieved R² values as low as 0.0374 in the Medium cluster. Feature importance analysis revealed that Inter-Arrival Time is the primary driver of bandwidth consumption, superseding raw payload volume, with direct implications for temporally-driven feature design. These results indicate that cluster-aware, leakage-free models offer a viable architecture for IoT bandwidth management, with cross-domain generalization as a primary direction for future work.
Evaluasi Sistem Rekomendasi Film Berbasis Sinopsis Cold-Start: Perbandingan TF-IDF, Word2Vec, dan FastText: Cold-Start Synopsis-Based Movie Recommendation: Comparing TF-IDF, Word2Vec, and FastText Holle, Khadijah Fahmi Hayati; Hidayah, Ela Ilmatul; Putri, Renata Amalia; Ningrum, Citra Ayu Putri
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 3 (2026): MALCOM July 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i3.2896

Abstract

Penelitian ini bertujuan untuk menentukan representasi teks ringan yang paling efektif dan efisien untuk rekomendasi film cold-start ketika sistem hanya memiliki sinopsis dan belum memiliki riwayat interaksi pengguna. Penelitian membandingkan TF-IDF, Word2Vec, dan FastText yang dikombinasikan dengan cosine similarity pada dataset film IMDb. Dari 10.001 data film, sebanyak 9.924 data valid digunakan setelah penghapusan sinopsis kosong, duplikasi, dan token tidak informatif. Pengujian dilakukan menggunakan 20 skenario kueri cold-start yang terdiri atas sepuluh kueri panjang bernarasi dan sepuluh kueri pendek berbasis kata kunci. Karena dataset tidak menyediakan label relevansi eksplisit dari pengguna, relevansi diukur menggunakan genre-proxy relevance. Evaluasi menggunakan Precision@10, HitRate@10, AP@10, MRR@10, nDCG@10, F1 skenario, serta waktu retrieval rata-rata. Hasil menunjukkan bahwa TF-IDF + cosine similarity memperoleh kinerja terbaik dengan Precision@10 sebesar 0,835, AP@10 sebesar 0,884, MRR@10 sebesar 0,942, nDCG@10 sebesar 0,942, dan F1 skenario sebesar 0,902. Kontribusi utama penelitian ini adalah pembuktian empiris terkontrol bahwa representasi leksikal dapat mengungguli rata-rata word embedding dalam rekomendasi cold-start berbasis sinopsis. TF-IDF lebih unggul karena sinopsis film yang pendek dan kueri tematik yang eksplisit masih mempertahankan kata kunci diskriminatif, sedangkan rata-rata vektor Word2Vec dan FastText cenderung meratakan sinyal kata kunci yang spesifik. Hasil ini menunjukkan bahwa TF-IDF tetap menjadi baseline yang kuat

Page 56 of 59 | Total Record : 581

 54   55   56   57   58 

Filter by Year

2021 2026


Reset
Filter By Issues
All Issue Vol. 6 No. 4 (2026): MALCOM October 2026 Vol. 6 No. 3 (2026): MALCOM July 2026 Vol. 6 No. 2 (2026): MALCOM April 2026 Vol. 6 No. 1 (2026): MALCOM January 2026 Vol. 5 No. 4 (2025): MALCOM October 2025 Vol. 5 No. 3 (2025): MALCOM July 2025 Vol. 5 No. 2 (2025): MALCOM April 2025 Vol. 5 No. 1 (2025): MALCOM January 2025 Vol. 4 No. 4 (2024): MALCOM October 2024 Vol. 4 No. 3 (2024): MALCOM July 2024 Vol. 4 No. 2 (2024): MALCOM April 2024 Vol. 4 No. 1 (2024): MALCOM January 2024 Vol. 3 No. 2 (2023): MALCOM October 2023 Vol. 3 No. 1 (2023): MALCOM April 2023 Vol. 2 No. 2 (2022): MALCOM October 2022 Vol. 2 No. 1 (2022): MALCOM April 2022 Vol. 1 No. 2 (2021): MALCOM October 2021 Vol. 1 No. 1 (2021): MALCOM April 2021 More Issue