cover
Contact Name
Mutammimul Ula
Contact Email
mutammimul@unimal.ac.id
Phone
+6281328661999
Journal Mail Official
jurnal.sisfo@unimal.ac.id
Editorial Address
Prodi Sistem Informasi Fakultas Teknik Universitas Malikussaleh Kampus Utama Cot Tengku Nie Reuleut Muara Batu, Aceh Utara, Provinsi Aceh, Indonesia Telp : +62.645.41373, Fax : +62.645.44450
Location
Kota lhokseumawe,
Aceh
INDONESIA
Sisfo: Jurnal Ilmiah Sistem Informasi
ISSN : 2598599X     EISSN : 25990330     DOI : https://doi.org/10.29103/sisfo.v6i1.7950
Jurnal Sistem Informasi Merupakan bidang keilmuan sistem informasi dan teknologi informasi dengan memuat artikel ilmiah penelitian murni dan terapan serta ulasan mengenai metode dan perkembangan teori, serta ilmu-ilmu terapan yang terkait dengan teknologi informasi serta informatika.Jurnal Sistem Informasi diterbitkan oleh Program Studi Sistem Informasi. Redaksi mengundang para peneliti, praktisi untuk menulis artikel ilmiah di bidang yang berkaitan dengan sistem informasi dan teknologi informasi serta informatika.Jurnal Sistem Informasi diterbitkan 2 (dua) kali dalam 1 tahun pada bulan Mei dan Oktober.
Articles 238 Documents
SAW-Based Optimization of Order Distribution on the SR12forLife Multi-Level Sales Platform Yori Adi Atma; J. Prayoga
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27048

Abstract

The rapid expansion of digital commerce has increased the need for efficient and fair product distribution systems, particularly in multi-level sales platforms where customer orders must be assigned to appropriate agents. Determining the best agent is a complex task because the decision should simultaneously consider delivery efficiency, stock availability, and balanced transaction opportunities among agents. This study proposes and evaluates the use of the Simple Additive Weighting (SAW) method to optimize product order distribution on the SR12forLife platform. The proposed model assesses candidate agents using four criteria: distance between customer and agent, available stock, number of previous received orders, and agent activity status. The study employed an operational test dataset consisting of 20 active agents located in several areas of West Sumatra and 105 customer transactions. Implementation and testing were conducted through a web-based system supported by Jupyter Notebook analysis. The experimental results confirmed that the proposed model performed effectively in allocating customer orders. All 105 transactions were successfully assigned to eligible agents, resulting in a 100% allocation success rate without stock shortages. Furthermore, all 20 active agents participated in the allocation process, indicating balanced workload distribution across the sales network. The model also achieved 92.38% regional suitability, where most customer orders were assigned to agents located in the same city or nearest area. These findings demonstrate that the SAW method is effective in producing objective, efficient, and fair order allocation decisions. These results indicate that the proposed model can improve both allocation fairness and operational efficiency in multi-level sales systems. Therefore, the proposed model can be considered an effective and scalable solution for optimizing order distribution in web-based multi-level sales platforms.
A Comparison Of Multiple Linear Regression And Single Exponential Smoothing For Predicting Rice Production In North Aceh Cindy Anindia Putri; Angga Pratama; Zalfie Ardian
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27049

Abstract

North Aceh Regency is the largest rice producer in Aceh Province; however, rice production has fluctuated and shown a downward trend from 2019 to 2023. This study aims to analyze the factors influencing rice production and to compare the Multiple Linear Regression and Single Exponential Smoothing (SES) methods in predicting rice production in North Aceh Regency. The variables used include harvested area, productivity, rainfall, and flooding. The Multiple Linear Regression method was used to analyze the relationships among variables, while SES was used to make predictions based on historical data. The analysis results indicate that productivity has the greatest influence on rice production, followed by harvested area. Rainfall and flooding also affect production, but their influence is relatively small. Model evaluation shows that Multiple Linear Regression has an MSE of 2156056.09, a MAD of 627, and a MAPE of 9.04%, which is better than Single Exponential Smoothing with an MSE of 51,613,195, a MAD of 3,867, and a MAPE of 28.57%. Based on these results, the Multiple Linear Regression method has a higher level of accuracy in predicting rice production in North Aceh.
Quality Analysis of Web-Based Visitor Management System (DATENG) Using Cypress Testing and Task-Based Usability Testing Methods Based on ISO 9241-11 (Case Study of PT Perta Arun Gas) Rizky Putra Fhonna; Yesy Afrillia; Ilham Sahputra; Sayed Fachrurrazi; Faiz Fadhilla
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27052

Abstract

Manual management of visitor data in the company environment has the potential to cause various problems, such as time inefficiency, the risk of data loss, and the occurrence of physical queues during certain operational hours. These conditions can hinder the smooth running of operational activities and reduce the quality of service to guests. Therefore, a web-based visitor management system called DATENG was developed which aims to support the process of recording, scheduling, and verifying visits in an integrated and structured manner. This research activity focuses on testing and evaluating the quality of the DATENG system to ensure that the system can function properly and is easy to use by users. The method used consists of two main approaches, namely functionality testing and usability evaluation. Functionality testing is carried out using the Cypress Testing method with an end-to-end testing approach to verify that all system features are running according to the needs that have been set. Furthermore, usability evaluation is carried out using a task-based testing method based on the ISO 9241-11 standard, which includes measuring aspects of effectiveness, efficiency, and satisfaction. The test results show that all the key features of the DATENG system can function properly without significant functional errors being found. Usability evaluations show that users are able to complete each task with a high success rate, relatively efficient turnaround time, and a good level of user satisfaction. Based on these results, it can be concluded that the DATENG system has good system quality and is able to support the visitor management process effectively, efficiently, and provide a positive user experience in accordance with the company's operational context.
Plant Disease Object Detection on PlantDoc Using YOLO26n Ovide Decroly Wisnu Ardhi; Rudi Hartono; Nanang Maulana Yoeseph
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27063

Abstract

Plant disease recognition from images is often reported as classification, although field inspection needs more than a class label. A farmer or agricultural officer must also know where the suspected leaf or disease region appears. This paper examines that localization problem using YOLO26n on the PlantDoc object detection dataset. PlantDoc is not a clean laboratory leaf dataset. It contains outdoor images with background clutter, uneven illumination, different leaf poses, overlapping objects, and visible symptom variation. YOLO26n was trained for 50 epochs with 416 × 416 input size and batch size 16. On the test set, the model obtained 0.534 precision, 0.560 recall, 0.547 F1-score, 0.573 mAP@0.50, and 0.417 mAP@0.50:0.95. Compared with the original PlantDoc detection benchmark, mAP@0.50 increased from 0.389 to 0.573. This result shows that a recent lightweight YOLO detector can improve object-level localization on PlantDoc. At the same time, the lower mAP@0.50:0.95 shows that precise bounding-box placement is still difficult. Most errors appear in visually similar symptoms, overlapping leaves, cluttered backgrounds, and under-represented classes. Thus, YOLO26n is better positioned as an initial baseline reference than as a deployable diagnostic model. Keywords: Object detection; Plant disease; PlantDoc; YOLO26n; Deep learning.
Integrated E-Learning System Development Using Next.js with Anti-Cheating Testing and Payment Integration Eva Yumami; Muhammad Ikhsan Wibowo; Agusviyanda; M. Khairul Anam
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27066

Abstract

The advancement of information technology has significantly increased the adoption of e-learning in the educational process. However, many e-learning systems still face limitations, such as the lack of integration between learning management, evaluation systems, and payment features. This study aims to design and implement a web-based e-learning system called NexusLearn that integrates course management, Computer-Based Testing (CBT) with anti-cheating mechanisms, and a payment gateway into a unified platform. The system is developed using the Software Development Life Cycle (SDLC) method, which includes requirement analysis, system design, implementation, and testing. The application is built using the Next.js framework and applies Role-Based Access Control (RBAC) to manage user access based on roles. The results show that the system successfully integrates learning, assessment, and transaction features into a single platform. The CBT module supports dynamic question navigation, real-time timers, automatic answer saving, and anti-cheating mechanisms such as tab-switch detection and automatic submission. The payment feature enables secure and automated transactions through Midtrans integration. Based on Black-box Testing results, all system functionalities operate as intended across key features, including authentication, course management, examination, and payment processing. These findings indicate that the NexusLearn system effectively improves integration, security, and efficiency in e-learning environments and has strong potential for implementation in educational institutions.
Comparative Analysis of SVM Accuracy and Effectiveness on Weka and Google Colab for Job Recommendations Rafika Sani; Nopa Wilyanita; Suparmi Suparmi; Tasya Maulidia
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27101

Abstract

Data Mining is a crucial technique for processing large and complex datasets to uncover hidden patterns that support strategic decision-making. This study evaluates the effectiveness of the Support Vector Machine (SVM) method implemented using Weka and Google Colab to classify employees into two categories: those recommended for promotion (positive class) and those not recommended (negative class). The research aims to compare the performance of both implementations in providing accurate and objective job recommendations. The dataset used in this study consists of employee performance evaluation records. The SVM model was trained and tested using different implementation platforms: Weka and Google Colab. In Weka, the model was configured with a polynomial kernel and achieved an accuracy of 99.62%, with precision, recall, and F-measure values of 1.000, 0.952, and 0.975 for the "recommended" class and 0.996, 1.000, and 0.998 for the "not recommended" class. Meanwhile, the implementation in Google Colab, using the LIBSVM library with a polynomial kernel, produced an accuracy of 97.85%, with precision, recall, and F-measure values of 0.984, 0.932, and 0.957 for the "recommended" class and 0.978, 0.986, and 0.982 for the "not recommended" class. The comparison results indicate that the Weka implementation provides slightly higher accuracy and better classification performance. However, Google Colab offers more flexibility and scalability, making it suitable for handling larger datasets. The findings of this study highlight the potential of SVM as a reliable tool for employee performance evaluation and job promotion recommendations. The use of machine learning in human resource management can enhance decision-making processes, ensuring fairness and efficiency in personnel assessments.
Diabetes Risk Prediction Using XGBoost-Based Machine Learning and Explainable Artificial Intelligence Sukasih Sukasih; Budi Arham; Nelti Rizka; Muhammad Badar
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27124

Abstract

Diabetes mellitus is a widespread chronic disease that affects populations worldwide and is often only identified after complications have occurred. Machine learning techniques can be applied to estimate the likelihood of diabetes based on clinical patient data. This research focuses on developing a predictive model using the Extreme Gradient Boosting (XGBoost) algorithm, along with the application of Explainable Artificial Intelligence (XAI) to interpret the model’s outcomes. The dataset used in this study consists of 768 patient records with eight clinical attributes, namely pregnancies, glucose level, blood pressure, skin thickness, insulin level, body mass index (BMI), diabetes pedigree function, and age. The research process includes data preprocessing, exploratory data analysis, model development, performance evaluation, and interpretation using SHAP. The findings indicate that the XGBoost model achieves high predictive performance and is capable of identifying key factors associated with diabetes risk. Based on SHAP interpretation, glucose level, BMI, age, and insulin are the most influential variables in the prediction process. The integration of machine learning and explainable AI improves model interpretability while maintaining reliable prediction performance.
Analysis of TikTok Acceptance on Generation Z’s Continuance Intention in Accessing Digital Information Using the UTAUT 3 Afifa Lutfia Fakhira; Dedy Setiawan
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27140

Abstract

This study aims to analyze acceptance of TikTok and its effect on continuance intention of Generation Z in obtaining digital information using UTAUT3 model, extended with Trust as an external variable. The increasing use of TikTok as a digital information source among Generation Z raises concerns regarding factors influencing user acceptance and continuance intention, which remain underexplored. This research examines effects of Performance Expectancy, Effort Expectancy, Social Influence, Facilitating Conditions, Hedonic Motivation, Habit, Personal Innovativeness, and Trust on Behavioral Intention and Continuance Intention. A quantitative approach was employed by distributing an online questionnaire to 165 Generation Z TikTok users in Indonesia. The data were analyzed using Structural Equation Modeling (SEM) with Partial Least Squares (PLS) method. Indicate Behavioral Intention significantly affects Continuance Intention. PE, EE, FC, and TR significantly influence BI, while FC, HB, PI, and TR significantly affect CI. In contrast, HB, HM, PI, and SI do not significantly affect BI. The insignificant result of Social Influence indicates social factors may not be the main determinant of Generation Z’s intention, although this aspect has not been explored in depth. This contributes to technology acceptance literature; however, it is limited to Generation Z users in Indonesia and uses a cross-sectional design, which limits generalizability and observation of behavioral changes.
IoT Based Fire Early Warning System Using ESP32 and Telegram with Multi Sensor Integration Budi Handoyo; Iqbal Iqbal
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27142

Abstract

Fire poses a severe threat, leading to significant loss of life, property, and environmental damage, underscoring the critical need for early detection and rapid response. This study proposes and implements an Internet of Things (IoT)-based fire early warning system utilizing an ESP32 microcontroller, integrated with multiple sensors, and Telegram as a real-time communication platform. The system continuously monitors environmental conditions through a DHT22 temperature sensor, an MQ-7 carbon monoxide (CO) gas sensor, and a flame sensor. A rule-based approach is employed to classify conditions as either normal or fire events, based on predefined threshold values, specifically temperature greater than or equal to 40 degrees Celsius and CO concentration greater than or equal to 200 ppm. To ensure comprehensive alerting, the system incorporates a dual-layer warning mechanism, providing local alerts via a buzzer and remote notifications through Telegram. An interactive Telegram bot interface is also implemented, facilitating real-time monitoring and multi-user notification management. Performance evaluation, conducted using a confusion matrix with 300 testing samples consisting of 150 normal and 150 fire conditions, demonstrated high classification efficacy. The system achieved an accuracy of 92.6 percent, a precision of 93.2 percent, a recall of 92.0 percent, and an F1-score of 92.6 percent. Furthermore, the system exhibited excellent responsiveness, with an average notification delay of 3.2 seconds, indicating near real-time performance. This integration of multi-sensor detection and Telegram-based communication significantly enhances the reliability and accessibility of fire alerts, offering an effective, low-cost, and scalable solution suitable for various early fire warning applications.
Analysis of Smartphone Addiction of Immanuel Medan Students Using Data Mining Classification Method (Naive Bayes and C4.5) Berna Susanti Br Tarigan; Resa Jesiulina H; Elvis Sastra Ompusunggu
Sisfo: Jurnal Ilmiah Sistem Informasi Vol. 10 No. 1 (2026): Sisfo: Jurnal Ilmiah Sistem Informasi, Mei 2026
Publisher : Universitas Malikussaleh

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29103/sisfo.v10i1.27184

Abstract

Smartphone addiction among students is a problem that interferes with their concentration while studying, social interactions, and their academic motivation. This study analyzed the level of smartphone addiction among students of SMKS Immanuel Medan using the Naive Bayes classification algorithm and the C4.5 decision tree. This study adopts a comparative quantitative approach using the phases of Knowledge Discovery in Databases (KDD), including data collection, data cleaning, data selection, data transformation, data mining, and evaluation. The research data was collected by distributing questionnaires to 100 students at SMKS Immanuel Medan. The study variables included age, gender, duration of smartphone use, purpose of smartphone use, dominant type of social media, and the level of smartphone addiction as target variables. The classification was carried out using RapidMiner software with a 70:30 training and testing data split. Model evaluation was carried out using a confusion matrix with the parameters of accuracy, precision, recall, and F1 score. The results show that the C4.5 decision tree algorithm gives better results than the Naive Bayes algorithm. The C4.5 algorithm achieved 90% accuracy, 88.9% precision, 80% recall, and 84% F1 score, while the Naive Bayes algorithm achieved 80% accuracy, 80% precision, 66.7% recall, and 73% F1 score. This research contributed to the development of a simple web-based expert system that helps schools assess the level of smartphone addiction among students quickly, objectively, and systematically, so that it can be used as a decision-making tool to monitor smartphone use in the education sector.