JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI

Comparison of Machine Learning Algorithm Performance on Healthcare and Technology Datasets

Grafita Elang, Muhammad Naufal; Meilany Nonsi Tentua — Mon, 22 Jun 2026 00:00:00 +0000

Machine Learning has become a powerful tool across different domains, including healthcare and consumer technology. This study compares the performance of classical Machine Learning algorithms—Support Vector Machine (SVM), Naive Bayes (NB), K-Nearest Neighbors (KNN), and Decision Tree (DT)—using two contrasting datasets: the Mobile Price Classification dataset and the Lung Cancer Survey dataset. Both datasets undergo identical preprocessing methods including normalization, label encoding, and dataset splitting (80:20 and 90:10). Evaluation metrics include accuracy, precision, recall, F1-score, and ROC-AUC. The results show that SVM provides the most consistent performance across both domains, while Decision Tree performs exceptionally well on the medical dataset but poorly on the technology dataset due to overfitting. This cross-domain experiment highlights how dataset characteristics significantly influence algorithm performance and provides insights for selecting appropriate algorithms based on domain-specific requirements.

Keywords: Classification, Lung Cancer, Machine Learning, Mobile Price, SVM,

Application of the Random Forest Algorithm for Stock Purchase Recommendation in Indonesian State-Owned Banks

Huzaeni Muhammad Hasan — Mon, 22 Jun 2026 00:00:00 +0000

The stock market of Indonesian state-owned enterprises (SOEs) presents significant investment potential but is characterized by high volatility and complex influencing factors, making it difficult for investors—especially beginners—to make informed decisions. This research proposes a comprehensive stock recommendation system based on the Random Forest algorithm, designed to analyze and predict SOE stock performance through the integration of historical price data, financial report indicators, and market sentiment analysis. The system aims to identify optimal stock purchase recommendations by capturing both quantitative and qualitative market dynamics. Model performance was assessed using the Mean Absolute Error (MAE) metric, demonstrating that the Random Forest-based approach effectively enhances prediction accuracy and reliability. The findings suggest that this model can serve as a valuable decision-support tool for investors, improving market insight, minimizing financial risk, and supporting more data-driven investment strategies.

- Internet of Things and Fog Computing for Remote Patient Monitoring in Cardiovascular Diseases and Cancer: A Literature Review

Mon, 22 Jun 2026 00:00:00 +0000

The development of the Internet of Things (IoT) has opened up enormous opportunities for the implementation of Remote Patient Monitoring (RPM) to improve the quality of monitoring patients with chronic diseases, particularly cardiovascular and cancer cases. This literature study aims to analyze the technical design of the IoT-RPM system, network architecture (cloud-fog), cardiovascular service process engineering, and a patient-centered RPM conceptual framework based on five recent scientific articles. The method used is a directed literature review of articles on ESP32-based patient monitoring prototypes and vital sensors, fog-IoT Platforms for RPM, cardiac RPM design through Business Process Re-engineering, general IoT-RPM implementation, and analysis of the RPM concept in cancer patients. The results of the study show that IoT-based monitoring systems are capable of providing real-time monitoring of vital signs and infusion conditions with adequate accuracy and notification times of only a few seconds, thereby potentially reducing delays in clinical action and the workload of nurses. The fog-IoT architecture has been proven to reduce latency and increase throughput compared to a pure cloud architecture, while enabling pre-processing and data filtering close to the source, which is important for large-scale implementation. At the process level, the implementation of RPM and integrated information systems in cardiac services increases service capacity and reduces working time and emergency response time. Conceptual analysis confirms that effective RPM must fulfill the attributes of continuous monitoring, technology integration, multidisciplinary collaboration, accessibility, and patient-centered care. In conclusion, IoT-RPM has great potential to realize more proactive, efficient, and patient-centered healthcare services, but still faces challenges related to richer data integration, security and privacy, infrastructure readiness, and the need for organizational and nursing practice changes.

Clusterization Of Productive Youth In Sirampog District Using The K-Means Method

M. Ali Sunan — Mon, 22 Jun 2026 00:00:00 +0000

Adolescence is a transition from childhood to adulthood where maturity does not mean only physical maturity, but also psychological self-maturity, the age limit in question is that the body and soul are mature and can carry out marriage to realize the goals of marriage properly without ending in divorce. There are many clustering methods in data mining, one of which is the K-Means method which is included in the partition-based method, where this method divides data into a number of non-hierarchical groups (partitions) that try to partition data into clusters so that data that has the same characteristics will be included in the same cluster. Cluster evaluation in the form of a validity test of the Davies Bouldin Index (DBI) in clustering marriage readiness for adolescents of productive age in Sirampog District The method used in this study is the experimental method. The object of research in this study is productive adolescents (20-29 years) in Sirampog District. The research was conducted from June 17, 2024 – August 18, 2024 (two months) The results of the validity test resulted in 4 clusters with details of Cluster 1 (cluster_0) as many as 8 people, cluster 2 (cluster_1) as many as 16 people, cluster 3 (cluster_2) as many as 17 people, cluster 1 (cluster_0) as many as 8 people with a total sample of 50 people. The research conducted proves that the performance of the K-Means Clustering Algorithm can be used to cluster adolescents of productive age ready to get married

Dynamic Clustering Analysis Using K-Means With Adaptive Centroid Optimization

Rudi anto — Mon, 22 Jun 2026 00:00:00 +0000

Clustering is a fundamental technique in data analysis that aims to group data based on similarity characteristics. The K-Means algorithm is widely used due to its simplicity and computational efficiency; however, it suffers from limitations related to static centroid initialization, which may lead to suboptimal local convergence. This study proposes a dynamic clustering approach based on adaptive centroid optimization to address these limitations by allowing centroid positions to evolve iteratively according to the data distribution. The proposed method incorporates an adaptive centroid update mechanism that considers inter-cluster distances and data density, enabling a more flexible and responsive clustering process. The evaluation is conducted using multiple datasets and assessed through performance metrics such as the Silhouette Score and Davies-Bouldin Index. The results demonstrate that the proposed approach improves cluster stability and reduces misclassification compared to the conventional K-Means algorithm. Furthermore, it shows better performance when handling datasets with uneven distributions. Therefore, adaptive centroid optimization offers a promising alternative for enhancing clustering quality in various data analysis applications.

Black Box Testing on Android-Based Augmented Reality Application of Human Digestive System

Novita Lia Nurilam Sari, Linda Wahyu Widianti — Mon, 22 Jun 2026 00:00:00 +0000

The development of digital technology in education has encouraged the use of more interactive learning media to improve students’ understanding of learning materials. One of the science subjects considered difficult for junior high school students is the human digestive system because the digestive process and organ structures cannot be observed directly. Based on these problems, this study aims to develop an Android-based Augmented Reality application for the Human Digestive System as an interactive learning medium and to implement the Black Box Testing method to evaluate the functionality of the application system. The software development method used in this research is Rapid Application Development (RAD) with marker-based tracking technology. This study applied a qualitative research method with data collection techniques consisting of observation, interviews, and literature study. Black Box Testing was conducted to determine the suitability of system functions based on input and output without analyzing the internal structure of the program code. The testing process involved 172 test scenarios covering marker scanning, 3D object visualization, rotation features, audio, video, learning materials, quizzes, navigation, and application exit functions. The results showed that all application features functioned properly according to expectations without any functional errors. The application was able to display 3D models of human digestive organs effectively and provide interactive features that support the learning process. Therefore, the Android-based Augmented Reality Human Digestive System application is considered feasible and effective as an interactive and educational learning medium.

Testing The Interface and Functionality of The Siakad Integrated E-Learning Features at Stmik Jakarta Sti&K With Black Box Testing

Muhammad Farhan, Linda Wahyu Widianti — Mon, 22 Jun 2026 00:00:00 +0000

The development of information technology in education has encouraged universities to implement digital learning systems integrated with the Academic Information System (SIAKAD). STMIK Jakarta STI&K utilizes e-Learning features as an online learning medium that supports material distribution, assignment submission, quizzes, and academic communication between lecturers and students. This study aims to test the interface and functionality of the integrated e-Learning feature within SIAKAD using the Black Box Testing method. Testing was conducted on the login, dashboard, material upload, assignment submission, online quizzes, menu navigation, and system logout features. Black Box Testing was used because it can evaluate system functions based on input and output without looking at the program code structure. The test results showed that all key system features functioned according to user requirements and produced outputs consistent with the test scenario. Furthermore, the system interface was deemed responsive, easy to understand, and supported user convenience in accessing online learning services. This study demonstrates the critical role of Black Box Testing in maintaining the quality of web-based information systems, particularly digital learning systems in higher education.

The Cybersecurity Awareness Education Activities In The Use Of Digital Technology For Junior And Senior High School Students

Mon, 22 Jun 2026 00:00:00 +0000

The community service activity entitled “Environmentally Friendly AI-Based Smart Computer Education for Students at Vihara Maha Wihara Dharmakirti” aims to improve students’ understanding of Artificial Intelligence (AI) technology while fostering awareness of wise and environmentally friendly technology usage. This activity is motivated by the still low level of students’ digital literacy, particularly in understanding the development of AI technology and the implementation of green computing concepts in daily life. The activity will be conducted offline on April 11, 2026, at Vihara Maha Wihara Dharmakirti, involving approximately 25–50 junior and senior high school students. The method used is an educational socialization approach with interactive activities, including material presentations, simple demonstrations of AI applications, discussions, and question-and-answer sessions to help participants better understand the material. The evaluation of the activity will be carried out through interactive discussions and question-and-answer sessions to measure participants’ level of understanding of the material presented. Through this activity, it is expected that students will not only understand the basic concepts of AI technology but also be able to apply them wisely, responsibly, and in an environmentally friendly manner in their daily lives, so that they can become a digitally intelligent generation that also cares about environmental sustainability.

DEVELOPMENT OF A WEB-BASED TOURIST ATTRACTION RECOMMENDATION SYSTEM IN BLITAR REGENCY USING THE WATERFALL METHOD

Cahyo Saputra, Daviq Cahyo Lukmanto, Putri Perwitasari, Indyah Hartami Santi — Mon, 22 Jun 2026 00:00:00 +0000

Tourism plays an important role in improving the regional economy, but the tourism potential in Blitar Regency has not been utilized optimally due to the lack of integrated and accurate information access. Tourists still face difficulties in obtaining information regarding locations, facilities, ticket prices, and access to tourist destinations because the information is scattered across various platforms. This study aims to design and develop a web-based tourist attraction recommendation system capable of providing information in a structured, easily accessible manner that aligns with user preferences. The importance of this study lies in its effort to support digital transformation in the tourism sector and to enhance user experience in selecting tourist destinations. The method used is the Waterfall method, which includes the stages of requirements analysis, system design, implementation, testing, and maintenance. Data collection was carried out through observation, interviews, and documentation. The system was developed using web technologies with a MySQL database and is equipped with recommendation features based on categories, ratings, and user reviews. The results show that the developed web-based tourism recommendation system has an easy-to-understand interface and functional website features that meet user needs. System testing was conducted with 21 respondents across 7 assessment aspects, with sample results indicating that the interface design aspect achieved an approval rate of 80.95%, while the system feasibility aspect achieved an approval rate of 85.71%. These results indicate that the system is considered effective, easy to use, and feasible to be implemented as a digital tourism information medium in Blitar Regency.

Testing the Reliability of Information Technology Infrastructure using the IT Capability Framework at SMK IPIEMS Surabaya

Mon, 22 Jun 2026 00:00:00 +0000

This research aims to implement the IT Capability Framework using the COBIT 2019 framework by evaluating the reliability of the information technology infrastructure at SMK IPIEMS Surabaya, with a focus on Deliver, Service, and Support (DSS). In today’s digital age, the role of IT infrastructure is crucial, particularly in this context to support academic and administrative processes. The study employed a quantitative approach, collecting data through structured interviews, observations, and document analysis involving IT staff and teachers. The research results classified the information technology system at SMK IPIEMS Surabaya have a cumulative score of 3.69 and a rating of “Good.” The domains of IT expertise, IT infrastructure, and IT management capabilities as a whole have achieved a “Good” rating. However, in addition to demonstrating that IT implementation has effectively supported school operations, the study also identified areas for future evaluation and improvement, such as expanding hardware resources, improving internet network quality, and implementing a more structured long-term IT strategic plan. These findings provide a data-driven foundation for enhancing IT governance and the performance of the institution.