SATESI: Jurnal Sains Teknologi dan Sistem Informasi

Perancangan dan Impelementasi Sistem Informasi Kepegawaian pada PT. Rezeki Bersama Energi

Defriel Lazuardi Mahendra — 2026-04-15

The rapid development of information technology requires organizations to manage human resource data effectively and efficiently in order to support organizational performance. PT Rezeki Bersama Energi currently manages its personnel administration manually using Microsoft Word and Microsoft Excel, which leads to several issues, including delays in administrative processes, a high risk of data entry errors, data duplication, and difficulties in retrieving and presenting personnel information. This study aims to design and implement a web-based Human Resource Information System to support integrated and structured personnel data management. The system development employs an iterative method, while system analysis is conducted using the PIECES framework to identify problems in the existing system. The developed system provides features for employee data management, attendance, leave requests, overtime, and employee rewards, with three levels of access rights: administrator, employee, and management. The results indicate that the implemented Human Resource Information System improves the efficiency and effectiveness of personnel administrative processes, minimizes data recording errors, and assists management in monitoring and making informed decisions related to human resource management in a more accurate and timely manner.

Analisis Sosio-Teknikal Disrupsi AI: Transformasi Arsitektur Pembelajaran dari Digital Assistance Menuju Human-Machine Co-Evolution di Pendidikan Vokasi

Dita Rahmawati — 2026-04-15

This study aims to analyze the impact of artificial intelligence (AI) disruption on the shift in learning paradigms in Indonesian higher education, particularly at Prasetiya Mandiri Polytechnic PSDKU Palembang. The global phenomenon demonstrates that AI has become a disruptive force, serving not only as an administrative tool but also as a collaborative partner in the teaching and learning process. This study employs a descriptive, qualitative approach, drawing on phenomenological methods, to understand the experiences of lecturers and students in their interactions with AI. Data was collected through in-depth interviews, observations of digital activities, and documentation studies from UNESCO, OECD, and Ministry of Education and Culture reports. The data analysis was carried out thematically, focusing on four main themes: digital assistance, changing learning patterns, human–AI collaboration, and ethical challenges in education. The results indicate that AI integration has enhanced learning efficiency, increased student participation, and fostered independent learning. However, negative impacts were also observed, including technology dependence, a decline in critical thinking skills, and the emergence of ethical dilemmas related to plagiarism and algorithmic bias. This research emphasizes that AI should be placed not as a substitute for educators, but as a collaborative partner that enriches the humanistic learning process. It is necessary to strengthen digital literacy, AI ethics, and an adaptive curriculum grounded in human-machine synergy so that educational transformation in the era of technological disruption can occur in a sustainable and equitable manner.

Analisis Faktor-Faktor Penyebab Depresi Mahasiswa di Indonesia Menggunakan Metode Regresi Logistik

Kiki Mustaqim — 2026-04-15

Depression is one of the most common mental health disorders experienced by university students and can have a serious impact on their psychological state, academic performance and social interactions. Academic pressure, financial demands, and changes in living environment are often factors that trigger an increased risk of depression in this age group. Therefore, a comprehensive analysis is needed to identify factors that contribute to the emergence of depression so that prevention efforts can be targeted. This study aims to analyze the factors associated with depression among university students in Indonesia using logistic regression method as a classification approach. The research data was obtained from the Kaggle platform and included several independent variables, namely age, gender, academic pressure, sleep duration, diet, financial stress, study satisfaction, and suicidal thoughts. The results of the analysis showed that the suicidal thoughts variable was the most significant factor affecting the likelihood of students experiencing depression, with a coefficient value of 15.0964. In addition, the logistic regression model built is able to provide good prediction performance with an accuracy rate of 95%. The findings are expected to serve as a basis for educational institutions and policy makers in designing early detection strategies, interventions, and depression prevention programs to create a healthier and more supportive campus environment.

Factors Influencing Customer Purchase Decisions in AI-Driven Online Shopping: Systematic Review

Arnold Aribowo — 2026-04-15

This paper offers a PRISMA-guided systematic literature review to examine how Artificial Intelligence (AI) is applied in e-commerce. The focus is on identifying key factors that influence customer purchasing decisions in AI-driven online transactions. It examines Information Systems (IS) theories relevant to the integration of AI and e-commerce, offering insights into frameworks used to analyze the relationship between AI and consumer behavior. Additionally, the paper identifies gaps in current research and provides recommendations for future studies, particularly in areas requiring further exploration to understand the evolving impact of AI on e-commerce. Through a review of existing literature, the study identifies critical factors such as perceived enjoyment, perceived usefulness, perceived ease of use, interactivity, consumer engagement, AI technology, and information quality, which significantly affect consumer purchase intentions. This review finds that Stimulus-Organism-Response (SOR) and Technology Acceptance Model (TAM) are the most commonly adopted theories, while Media Richness Theory is used less frequently. The findings provide a robust foundation for future research, enabling the formulation of empirically testable hypotheses. Furthermore, this study offers a more integrated perspective by organizing identified constructs into a multi-dimensional framework and suggests directions for future empirical research, such as developing research models and validating them through survey-based approaches and Structural Equation Modeling (SEM-PLS), as well as qualitative methods. The study aims to offer insights to AI developers and e-commerce practitioners, helping them enhance AI-powered systems to better meet consumer needs and expectations, ultimately improving customer satisfaction and increasing purchase rates.

Peran Generative Artificial Intelligence dalam Meningkatkan Efisiensi Proses Pembelajaran di Tingkat Tinggi

Wincent Wisely — 2026-04-15

The development of Generative Artificial Intelligence (Gen AI) has significantly transformed the learning process, particularly in improving task completion efficiency. This study aims to analyze the effect of Gen AI usage on learning efficiency among high school students and university students. A quantitative approach was employed using a survey of 83 respondents. The variables examined include frequency of use, duration of use, and learning efficiency, which is measured based on task completion time. The results indicate that the level of Gen AI usage is relatively high, with a mean frequency of 3.51 and a mean duration of 3.53, while efficiency shows the highest mean value of 3.94. Regression analysis reveals that the model is statistically significant (p-value < 0.05) with a coefficient of determination of , indicating that 61.18% of the variance in learning efficiency is explained by the model. Partially, frequency of use has a positive and significant effect on efficiency ( ; p-value < 0.05), whereas duration of use is not statistically significant ( ; p-value > 0.05). These findings suggest that usage intensity plays a more critical role than usage duration. Overall, Gen AI is shown to enhance learning efficiency; however, its effectiveness depends on how users actively and appropriately engage with the technology.

Analisis Kinerja AES dan RC6 pada File Multimedia

Sugiyatno Sugiyatno — 2026-04-15

This study aims to analyze and compare the performance of symmetric cryptographic algorithms, namely Advanced Encryption Standard (AES) and Rivest Cipher 6 (RC6), in securing multimedia files. The research employs a quantitative experimental approach using various file formats, including JPG, PNG, MP3, and MP4, with different file sizes. Performance evaluation is conducted based on three main parameters: encryption time, decryption time, and throughput. The implementation is carried out in a controlled local environment using Python to ensure consistency and accuracy of measurements. The results show that AES consistently outperforms RC6 in terms of faster encryption and decryption processes as well as higher and more stable throughput across all tested formats and file sizes. In contrast, RC6 exhibits higher computational overhead due to its complex arithmetic operations, resulting in slower processing time and less stable performance. Furthermore, the findings indicate that file size and format significantly influence algorithm performance, where larger and more complex multimedia data require higher processing time. This study contributes a comprehensive multi-format evaluation framework that provides practical insights for selecting efficient cryptographic algorithms in real-world multimedia security applications.

Perbandingan Metode Naïve Bayes dan Support Vector Machine dalam Analisis Sentimen Citra Polri berdasarkan Opini pada Platform Twitter/X

Sophy Awaliah — 2026-04-15

The Indonesian National Police (Polri) is a law enforcement agency responsible for maintaining security and order in Indonesia. In the digital era, Polri’s image has increasingly been highlighted on social media platforms such as Twitter/X, which serve as a major channel for the public to express opinions and criticism. This study aims to compare the performance of the Naive Bayes method and Support Vector Machine (SVM) in sentiment analysis of public opinion toward Polri. Naive Bayes, known for its probabilistic approach, is compared with SVM, a robust machine learning algorithm capable of classifying data with clear margins between classes. The dataset was divided into 80% training data and 20% testing data with stratification to ensure balanced sentiment proportions. Performance evaluation was conducted using accuracy, precision, recall, and F1-score through a confusion matrix. Results show that SVM achieved the highest accuracy of 90%, while Naive Bayes obtained 83%. In terms of F1-score, SVM reached a macro average of 0.90 with its best performance in the positive category (0.97), while Naive Bayes reached 0.83 with its best in the positive category (0.90). Overall, SVM outperformed Naive Bayes, particularly in classifying neutral sentiment. This study provides insights into the effectiveness of SVM for analyzing informal tweets and can serve as a reference for future research and public opinion monitoring system development.