cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Fransiskus Panca Juniawan
Contact Email
Fransiskus Panca Juniawan
Phone
-
Journal Mail Official
fransiskus.pj@atmaluhur.ac.id
Editorial Address
-
Location
Kota pangkal pinang,
Kepulauan bangka belitung
INDONESIA
Jurnal Sisfokom (Sistem Informasi dan Komputer)
Published by Sekolah Tinggi Manajemen Informatika dan Komputer Atma Luhur
ISSN : 23017988     EISSN : 25810588     DOI : -
Core Subject : Science, Engineering,
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management
Jurnal Sisfokom merupakan singkatan dari Jurnal Sistem Informasi dan Komputer. Jurnal ini merupakan kolaborasi antara sivitas akademika STMIK Atma Luhur dengan perguruan tinggi maupun universitas di Indonesia. Jurnal ini berisi artikel ilmiah dari peneliti, akademisi, serta para pemerhati TI. Jurnal Sisfokom diterbitkan 2 kali dalam setahun yaitu pada bulan Maret dan September. Jurnal ini menyajikan makalah dalam bidang ilmu sistem informasi dan komputer.
Arjuna Subject : -
Articles 709 Documents
 67   68   69   70   71 
A Comparative Analysis Of K – Means And K – Medoids For Smartphone Clustering M. Arif Fadillah; Kurniabudi; Dodo Zaenal Abidin
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2660

Abstract

The rapid growth of the smartphone industry has increased market complexity, making objective segmentation more challenging. The objective of this study is to conduct a comparative evaluation of the K - Means and K - Medoids methods in grouping smartphone sales data based on multiple attributes, namely memory, storage, rating, selling price, and original price. Unlike prior studies, this research conducts a direct comparison using the same dataset and multiple evaluation metrics. A systematic data mining approach is implemented through the CRISP-DM framework, covering the stages of data understanding, preprocessing, modeling, and evaluation. The dataset comprises 3,114 smartphone instances, which are grouped into three clusters (k = 3), with performance measured using the Silhouette Coefficient and Davies–Bouldin Index (DBI). Based on the evaluation metrics, K - Medoids exhibits superior cluster cohesion with a Silhouette score of 0.344, exceeding that of K - Means (0.313). Conversely, K - Means demonstrates slightly better separation, as shown by its lower Davies–Bouldin Index (1.061 versus 1.079). Even so, K - Medoids is generally preferred due to its resilience to outliers and its consistency in producing stable clustering outcomes. These findings provide insights to support data-driven decision-making in smartphone market segmentation.
Student Reading Interest Segmentation in the Digital Era Using the K-Means Algorithm (Case Study: ISB Atma Luhur) Delpiah Wahyuningsih; Rezky Yuranda; Chandra Kirana; Anisah
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2674

Abstract

The rapid advancement of information technology in the digital era has triggered transformative changes in students’ reading behavior. Although access to information has become virtually unlimited, students are increasingly susceptible to digital distractions and tend to prefer instant content consumption over in-depth engagement with academic literature. This study aims to segment students’ reading interests at the Institut Sains dan Bisnis (ISB) Atma Luhur in order to obtain a more granular understanding of their literacy profiles through a data mining approach. The unsupervised learning algorithm K-Means Clustering was employed to group reading behavior patterns. Primary data were collected from 130 student respondents using a questionnaire instrument covering reading intensity, genre preferences, and motivation. The data preprocessing stage involved several steps, including data cleaning, data transformation, and feature selection, focusing on two quantitative attributes: the number of books read per month and daily reading duration. The determination of the optimal number of clusters was evaluated using the Elbow Method, which identified three as the optimal number of clusters. The clustering results revealed three dominant persona segments: incidental readers (34.62%), pragmatic academic readers (37.69%), and digital recreational readers (27.69%). These analytical findings provide strategic insights for institutional stakeholders and library managers in designing adaptive, personalized, and targeted literacy intervention programs aligned with the unique characteristics of each student segment within the digital ecosystem.
Bitcoin Closing Price Prediction Using LSTM with Google Trends Feature Engineering and Feature Selection Dionisius Nusaca Redegnosis Nolejanduma; Majid Rahardi
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

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

Abstract

Bitcoin price prediction has become an important research topic due to the high volatility and dynamic behavior of cryptocurrency markets. Traditional forecasting approaches often struggle to capture both complex temporal patterns and external market sentiment simultaneously. This study aims to predict the daily closing price of Bitcoin using a Long Short-Term Memory (LSTM) model integrated with Google Trends data as an external sentiment variable. The dataset consists of 1,926 daily records from December 30, 2020, to April 8, 2026, containing Bitcoin historical prices, technical indicators, and Google Trends features. The proposed methodology applies preprocessing, MinMax normalization, technical indicator generation, feature engineering, and Walk-Forward Validation using TimeSeriesSplit with 5 splits to avoid data leakage in time-series forecasting. Three models, namely XGBoost, LSTM, and Hybrid LSTM + XGBoost, were compared to determine the best baseline model. Experimental results show that the LSTM model achieved the best performance, with an average RMSE of 2053.25, outperforming both the XGBoost and the Hybrid LSTM + XGBoost models. However, integrating all Google Trends features decreased prediction performance due to increased noise. Feature selection on relevant Google Trends variables, such as the crypto market and BTC, successfully improved the model’s performance, although it still did not surpass the baseline LSTM model. The study concludes that LSTM is highly effective for Bitcoin price forecasting and that proper feature selection is essential when integrating external sentiment data, such as Google Trends.
An Ensemble Deep Learning Framework for Early Stunting Detection in Toddlers: Supporting the Free Nutritious Meal Program Suhartini Suhartini; Andi Christian; Iwan Setiawan
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2681

Abstract

Chronic malnutrition known as stunting remains a pressing public health concern in Indonesia, with a national prevalence of 19.8% reported in the 2024 Indonesian Nutrition Status Survey, still well above the 14.2% target set in the 2024–2029 National Medium-Term Development Plan. The Free Nutritious Meal Program (Makan Bergizi Gratis, MBG), launched in January 2025, requires a precise screening mechanism to ensure that nutritional interventions reach the right recipients. The present study is framed as a cross-sectional classification task—rather than a longitudinal predictive model—designed to support point-of-care screening at community health posts. This study proposes an ensemble deep learning framework for early stunting detection that combines three complementary learning paradigms: Random Forest (bagging), XGBoost (gradient boosting), and a Deep Neural Network (DNN). A publicly available Kaggle dataset of 120,999 toddler anthropometric records, in which class labels are likely derived deterministically from the WHO Height-for-Age Z-score (HAZ) formula, was used as the experimental basis. The pipeline included feature engineering grounded in WHO Child Growth Standards, class balancing via SMOTE applied exclusively to the training set, hyperparameter optimization using Optuna, and soft-voting integration of the three base learners. Evaluation was performed on a stratified test set (n=24,200) using accuracy, precision, recall, F1-score, and AUC-ROC, complemented by 10-fold cross-validation and SHAP-based interpretability analysis. The ensemble model achieved 99.84% accuracy, 0.9984 F1-score, and 1.0000 AUC-ROC, exceeding the proposal targets of 88% accuracy and 0.90 AUC-ROC. These near-perfect metrics should be interpreted as evidence that the model has successfully recovered the rule-based labeling structure of the dataset; performance on real-world Posyandu data—where biological variability, measurement error, and unobserved socioeconomic determinants are present—is expected to be lower, and external validation is therefore prioritized as future work. SHAP analysis identified the approximate Height-for-Age Z-score, height, and the height-by-gender interaction as the three most influential features, consistent with WHO anthropometric principles. These findings provide a technical foundation for AI-based screening systems deployable in community health posts and primary care clinics, supporting the effectiveness of the MBG program toward Indonesia Emas 2045.
Analysis of Smart City Quick Win Program Implementation Using Fuzzy BWM and TOPSIS in Manado and Tomohon Puteri Justia Kardia Momuat Wahani; Sri Yulianto Joko Prasetyo; Indrastanti R. Widiasari; Johan J. C. Tambotoh
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2690

Abstract

This study analyzes the implementation quality of smart city quick win programs in Manado and Tomohon using the Fuzzy Best-Worst Method (BWM) and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS). The evaluation covers 12 quick win programs across six smart city dimensions: smart governance, smart branding, smart economy, smart living, smart society, and smart environment. The assessment uses 12 criteria derived from government smart city evaluation elements, including policy alignment, problem and goal clarity, public benefit, implementation readiness, technological readiness, governance ownership, service performance, monitoring and evaluation, and sustainability. Data were collected from official documents, evidence catalogs, interviews, expert assessments from 20 respondents, and government evaluation records. The novelty of this study lies in the development of a program-level evaluation model that integrates Fuzzy BWM for criteria weighting and TOPSIS for ranking quick win programs. Unlike previous studies that mainly evaluate smart city performance at the city or dimension level, this study positions quick win programs as concrete implementation units. The results show that problem and goal clarity, direct public benefit, and service performance are the most influential aspects in determining implementation quality. The TOPSIS results identify Manado 360 and SmartGov/PONTER as the strongest programs, while the city-level comparison shows that Manado has more consistent implementation quality than Tomohon. This study contributes a structured decision-support model for evaluating and improving smart city quick win programs.
DigestAR: Android Based Augmented Reality Application for IPAS Digestive System Learning with the Fisher Yates Algorithm Muhammad Hasan Fikri; Febrian Dewanto; Nur Latifah Dwi Mutiara Sari
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2702

Abstract

The teaching of Natural and Social Sciences (IPAS) regarding the human digestive system in elementary schools still largely relies on two-dimensional media, causing students to struggle with understanding the concrete forms and functions of digestive organs. The purpose of this study is to develop an Android-based learning application called DigestAR that integrates Augmented Reality (AR) technology and Kuis (quiz) features to support the learning process for material on the human digestive system. The Multimedia Development Life Cycle (MDLC) approach was used to create the application, which in this study was limited to the testing phase, encompassing concept, design, content collection, assembly, and testing. The application was built using Unity with the C# programming language, the Vuforia SDK to support AR features, and the Fisher Yates algorithm to randomly shuffle quiz question order evenly, ensuring a different sequence in each session. The application’s main features include 3D visualizations of digestive organs, learning materials, and evaluation quiz. System testing involved Black Box Testing, White Box Testing, and User Acceptance Testing (UAT). The test results showed that all application functions operated as intended. The UAT results, which involved students and teachers at SDN Karangtempel in Semarang, as well as university students, yielded an average score of 96.67%, categorized as “very good.” These results indicate that the developed application has a high level of user acceptance and can serve as an alternative interactive learning medium to aid students’ understanding of the human digestive system.
CRISP-DM Based Sentiment Analysis on MSME Loan Opinions in Bangka Belitung Using Naïve Bayes Ari Amir Alkodri; Fitriyani Fitriyani; Melati Suci Mayasari; Yuyi Andrika; Sarwindah Sarwindah; Agus Dendi
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2711

Abstract

The development of the MSMEs sector plays a crucial role in national economic growth. It not only supports the regional economy but also significantly impacts and contributes to job creation and equitable income distribution. However, one of the primary obstacles faced by MSMEs is limited access to financing or loans. To address this issue, many government and private institutions provide financing and mentoring programs. This study focuses on the analysis of sentiment opinions regarding assisted MSMEs loans in the Bangka Belitung Islands Province using the Cross-Industry Standard Process for Data Mining approach and the Multinomial Naïve Bayes algorithm, was utilized for opinion sentiment analysis on assisted MSME loans, with a total of 1,112 reviews collected through surveys and data from assisted MSMEs, such as Witel. This study successfully implemented the CRISP-DM framework and the Multinomial Naïve Bayes algorithm to analyze public opinion sentiment toward assisted MSME loan programs in the Bangka Belitung Islands Province. Achieving an accuracy of 96.02%, this model proves to be highly effective and efficient in extracting and classifying survey-based opinion data. The primary scientific contribution of this research is the successful integration of a structured data mining approach with local economic policy analysis. However, a trade-off was identified in the Negative Recall of 0.79, indicating that 21% of negative opinions were missed due to a class imbalance where positive opinion data significantly outnumbered negative opinions in the survey. Overall, this approach yielded exceptionally high evaluation metrics, achieving a Positive Recall of 1.00 and a Negative Precision of 1.00.
Comparative Evaluation of Boosting Ensemble Models for Medication Adherence Prediction in Patients with Non-Communicable Diseases Ihya' Nashirudin Abrar; Muhammad Kunta Biddinika; Herman Yuliansyah
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2712

Abstract

Hypertension and diabetes mellitus are among the leading drivers of premature mortality worldwide. Long-term disease management depends critically on patient adherence to prescribed regimens; however, adherence rates in chronic-illness populations remain persistently low, particularly in developing regions. Although predictive studies on medication adherence have frequently employed Random Forest, Logistic Regression, and Support Vector Machines, a systematic benchmark of modern boosting ensembles on imbalanced clinical datasets has yet to be established. To address this gap, the present study evaluates five boosting algorithms — XGBoost, AdaBoost, Gradient Boosting, LightGBM, and CatBoost — using a publicly accessible medical claims dataset from the Cimas Medical Aid Society, Zimbabwe, comprising 24,084 patient records and 11 predictor variables. The dataset exhibits moderate class imbalance (59.85% non-adherent; 40.15% adherent). The experimental pipeline included data cleaning, stratified 80:20 splitting, class-weight calibration, uniform baseline hyperparameters (n_estimators = 100, learning_rate = 0.1), 10-fold stratified cross-validation, and Wilcoxon signed-rank statistical testing. LightGBM outperformed all competing models, achieving an accuracy of 0.8163, AUC-ROC of 0.9044, F1-scores of 0.8007 (adherent) and 0.8296 (non-adherent), and a Matthews Correlation Coefficient of 0.6540, with cross-validation confirming stability (0.8147 ± 0.0069). Feature importance analysis identified Annual Claim Amount, Units Total, and Age as the most informative predictors. This work delivers the first empirical benchmark of five contemporary boosting ensembles for NCD medication adherence prediction, integrating class-weighted training and statistical validation within a unified framework, offering actionable guidance for model selection in resource-limited clinical settings.
Explainable Machine Learning Framework for Outbound IoT Botnet DDoS Detection Muhammad David Fawwas; Bambang Agus Herlambang; Nur Latifah Dwi Mutiara Sari
Jurnal Sisfokom (Sistem Informasi dan Komputer) Vol. 15 No. 3 (2026): JULY
Publisher : ISB Atma Luhur

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.32736/sisfokom.v15i3.2715

Abstract

The swift expansion of Internet of Things (IoT) technology has markedly heightened cybersecurity threats, especially Distributed Denial-of-Service (DDoS) assaults executed via IoT botnets. Conventional Intrusion Detection Systems (IDS) predominantly concentrate on the examination of inbound traffic, limiting their capability to proactively identify compromised internal devices through abnormal outbound communication patterns. To address this gap, This paper presents an explainable machine learning architecture that uniquely combines outbound traffic analysis with SHapley Additive exPlanations (SHAP) based interpretability for IoT botnet DDoS detection  a combination not simultaneously addressed in prior work. The framework was evaluated using the IoT-23 dataset comprising four IoT botnet scenarios and 50,000 network flow records. Random Forest, XGBoost, and ensemble models were assessed using Accuracy, Precision, Recall, F1-Score, AUC-ROC, and Stratified K-Fold Cross Validation. Experimental results demonstrated that Random Forest achieved the best performance with approximately 97.5% accuracy, while XGBoost and ensemble models also produced consistently high classification results. SHAP analysis further identified orig_ip_bytes, proto_state_interaction, and orig_pkts_per_sec as the most influential indicators of IoT botnet outbound traffic behavior. Unlike previous studies that target inbound traffic or omit explainability, the proposed framework provides both early detection of compromised IoT devices and transparent decision support for cybersecurity analysts, making it more operationally relevant for real-world deployment. Nevertheless, this study remains limited to offline evaluation using a single benchmark dataset, suggesting the need for future real-time implementation and broader network environment evaluation.

Page 71 of 71 | Total Record : 709

 67   68   69   70   71 

Filter by Year

2013 2026


Reset
Filter By Issues
All Issue Vol. 15 No. 02 (2026): MAY Vol. 15 No. 01 (2026): JANUARY Vol. 15 No. 3 (2026): JULY Vol. 14 No. 4 (2025): NOVEMBER Vol. 14 No. 3 (2025): JULY Vol. 14 No. 2 (2025): MEY Vol. 14 No. 1 (2025): JANUARY Vol. 13 No. 3 (2024): NOVEMBER Vol 13, No 2 (2024): JULY Vol. 13 No. 2 (2024): JULY Vol. 13 No. 1 (2024): MARET Vol 13, No 1 (2024): MARET Vol 12, No 3 (2023): NOVEMBER Vol. 12 No. 3 (2023): NOVEMBER Vol 12, No 2 (2023): JULI Vol. 12 No. 2 (2023): JULI Vol 12, No 1 (2023): MARET Vol 11, No 3 (2022): NOVEMBER Vol 11, No 2 (2022): JULI Vol 11, No 1 (2022): MARET Vol 10, No 3 (2021): NOVEMBER Vol 10, No 2 (2021): JULI Vol 10, No 1 (2021): MARCH Vol 9, No 3 (2020): NOVEMBER0 Vol 9, No 2 (2020): JULI0 Vol 9, No 1 (2020): MARET0 Vol 9, No 3 (2020): NOVEMBER Vol 9, No 2 (2020): JULI Vol 9, No 1 (2020): MARET Vol 8, No 2 (2019): SEPTEMBER0 Vol 8, No 2 (2019): SEPTEMBER Vol 8, No 1 (2019): MARET Vol 7, No 2 (2018): September Vol 7, No 1 (2018): Maret Vol 6, No 2 (2017): September Vol 6, No 1 (2017): Maret Vol 5, No 2 (2016): September Vol 5, No 1 (2016): Maret Vol 4, No 2 (2015): September Vol 4, No 1 (2015): Maret Vol 3, No 2 (2014): September Vol 3, No 1 (2014): Maret Vol 2, No 1 (2013): Maret 2013 Vol 2, No 2 (2013): September More Issue