Garuda - Garba Rujukan Digital

Pendekatan Naive Bayes dalam Analisis Sentimen pada Ulasan Pengguna Aplikasi Indodax di Platform Google Play Store Riki Ardi Pranata; Susanto Susanto; Nur Wakhidah
Smart Comp :Jurnalnya Orang Pintar Komputer Vol 15, No 2 (2026): Smart Comp: Jurnalnya Orang Pintar Komputer
Publisher : Politeknik Harapan Bersama

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30591/smartcomp.v15i2.9938

Aplikasi investasi digital, seperti Indodax, berperan sebagai sarana transaksi jual beli aset kripto yang mendukung aktivitas pengguna sesuai dengan tujuan dan kebutuhan investasi. Analisis sentimen digunakan untuk mengidentifikasi opini serta kecenderungan sikap pengguna terhadap suatu topik. Penelitian ini bertujuan untuk menganalisis sentimen ulasan pengguna Aplikasi Indodax yang diperoleh dari platform Google Play Store. Metodologi yang diterapkan adalah Knowledge Discovery in Database (KDD), yang meliputi tahapan data selection, preprocessing, pelabelan berbasis lexicon-based, transformation, klasifikasi menggunakan algoritma Naive Bayes, serta evaluasi. Proses klasifikasi dilakukan untuk mengelompokkan ulasan ke dalam dua kategori sentimen, yaitu positif dan negatif. Dataset penelitian berasal dari ulasan pengguna Play Store yang telah melalui tahap prapemrosesan teks. Hasil pengujian menunjukkan algoritma Naive Bayes memberikan performa klasifikasi yang cukup baik. Berdasarkan tiga skenario pembagian data latih dan data uji, yaitu rasio 60:40, 70:30, dan 80:20, diperoleh rasio 60:40 menghasilkan kinerja optimal dengan nilai akurasi sebesar 82,95% serta nilai presisi, recall, dan F1-score sebesar 83%. Distribusi sentimen menunjukkan 55,45% ulasan bersifat negatif dan 44,55% bersifat positif, menandakan tanggapan pengguna terhadap aplikasi Indodax masih didominasi oleh sentimen negatif. Namun, metode pelabelan berbasis lexicon-based masih kesulitan dalam konteks kalimat seperti sarkasme, bahasa informal, dan ambigu dari sebuah ulasan.

Perancangan Aplikasi Pemantauan Aktivitas Fisik Mobile Berbasis User-Centered Design Purbo Pangestu Satria Jati; Susanto Susanto; Titis Handayani
Jurnal Teknologi Informasi dan Multimedia Vol. 7 No. 4 (2025): November
Publisher : Sekawan Institut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35746/jtim.v7i4.811

Physical activity monitoring is an important aspect of maintaining health and fitness. With increasing awareness of the importance of a healthy lifestyle, many people are trying to be more physically active. Although many health apps offer physical activity monitoring features, not all apps are designed with user comfort and needs in mind. This study aims to design and evaluate a prototype of a mobile-based health app user interface that emphasizes comfort in physical activity monitoring, using a User-Centered Design (UCD) approach. This approach places the user at the center of the entire design process, ensuring that the design aligns with users' actual preferences and needs. The research methodology includes stages of understanding the user context, identifying user needs, designing the interface (wireframes and user flow), and evaluating the design through A/B testing conducted internally or preliminarily, without involving external respondents. The results of the study indicate that design version B is superior to version A in terms of ease of use and user engagement, as evidenced by a 67.86% increase in interaction and a 71.43% increase in feature usage, based on quantitative metrics such as task completion count and average interaction time measured through internal task scenario simulations. Version B features a more modern appearance and simple, clear navigation. These findings underscore the importance of applying UCD principles in the development of effective and efficient health application interfaces. A better design can encourage user engagement in monitoring physical activity. The application, once designed, not only meets current user needs but also has a strong foundation for future development requirements.

Implementasi Model Machine Learning untuk Deteksi Phishing dengan Pendekatan Ekstraksi Fitur yang Dioptimalkan Adam Pradana; Susanto Susanto
Jurnal Teknologi Informasi dan Multimedia Vol. 8 No. 1 (2026): February
Publisher : Sekawan Institut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35746/jtim.v8i1.881

Phishing is a common form of cybercrime used by digital criminals to steal sensitive information such as passwords, personal data, and financial details through fake websites designed to re-semble legitimate pages. However, conventional detection methods such as blacklists and manual inspection are currently considered ineffective due to their static nature, often failing to recognize new, evolving and increasingly sophisticated attack patterns. To address this issue, this study developed a machine learning-based phishing detection model focused on improving the accura-cy and efficiency of identifying malicious sites. This model applies an optimized feature extrac-tion technique to enable the system to analyze URL characteristic patterns more comprehensively and targeted. The research dataset was taken from the Kaggle platform, which provides a dataset of phishing and benign URLs with a high reputation. The data was then processed through nor-malization, cleaning, and extraction of important features such as URL structure and domain at-tributes. The classification process was carried out using an ensemble learning approach that combines four popular algorithms: Random Forest, Gradient Boosting, Logistic Regression, and AdaBoost through a soft voting mechanism. The evaluation results show that the proposed model has excellent performance with an accuracy of 98.10%, a precision of 97.81%, a recall of 93.90%, an F1-Score of 95.82%, and a ROC-AUC of 98.62%. These findings confirm that the ensemble ap-proach with optimized features has great potential for application in artificial intelligence-based cybersecurity systems capable of adaptive and real-time phishing detection.

Perancangan dan Implementasi Sistem Pengelolaan Order Barang Berbasis Web dengan Laravel Nabela Yulian Anggraini; Susanto Susanto
Jurnal Teknologi Informasi dan Multimedia Vol. 8 No. 1 (2026): February
Publisher : Sekawan Institut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35746/jtim.v8i1.890

CV. Surya Sarana Dinamika currently faces significant operational inefficiencies due to its man-ual, paper-based order management process. These challenges lead to frequent data recording errors, substantial delays in information flow, and a critical lack of real-time visibility regarding order status, which ultimately hinders effective administrative decision-making. This study aims to address these systemic issues by designing and implementing a web-based order management system utilizing the Laravel Framework. The system was developed using the Agile Development methodology, employing iterative cycles to ensure that the final product remains aligned with dynamic business requirements. The architecture is built upon the Model-View-Controller (MVC) pattern to ensure code scalability and long-term maintenance ease. Functional validation via Black-Box testing confirms that all core features, including order submission and purchasing validation, operate according to specifications. Quantitatively, the implementation resulted in a drastic efficiency increase; the average time for order submission decreased from 20-25 minutes to just 3-5 minutes (an efficiency gain of over 75%), while purchasing validation time was re-duced from 30-60 minutes to 5-10 minutes. These findings demonstrate that digital automation effectively eliminates manual bottlenecks and reduces the risk of human error. By providing cen-tralized data and real-time tracking, the system significantly enhances accountability and trans-parency within the company’s supply chain, providing a replicable model for digital transfor-mation in similar organizational contexts..

Rancang Bangun Sistem Informasi Pendaftaran Siswa Baru Berbasis Web dengan Metode Waterfall pada TK Pertiwi 14 Gilang Akbar Romadhoni; Susanto Susanto; Titis Handayani
Jurnal Teknologi Informasi dan Multimedia Vol. 8 No. 1 (2026): February
Publisher : Sekawan Institut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35746/jtim.v8i1.895

The transformation of information technology in the modern era provides significant opportunities to improve the effectiveness of educational administration, particularly in the new student admission process. One form of implementation is a web-based registration system that can automate the online registration flow and minimize data recording errors. Pertiwi 14 Kindergarten currently still uses the conventional registration method through paper forms that require parents to be present in person at the school, potentially causing verification delays, information duplication, and data input errors. This study aims to design and build a web-based new student registration system using the Laravel framework with Laravel Filament support to display a dynamic interface and structured data management. The system development uses the Waterfall model through the stages of needs analysis, design, implementation, testing, and maintenance. Testing results using Black Box Testing show that all system functions run according to specifications without any errors and are able to reduce input errors by up to 100% compared to manual methods. In addition, the registration data verification process is faster with an increase in time efficiency of 70%. The implementation of this system makes a real contribution to improving ease of access, data accuracy, management transparency, and administrative efficiency in the new student admission process at Pertiwi 14 Kindergarten.

Implementasi Algoritma FP-Growth untuk Sistem Rekomendasi Produk Kebutuhan Pokok pada E-Commerce Mahjid Herlambang; Susanto Susanto
Jurnal Teknologi Informasi dan Multimedia Vol. 8 No. 1 (2026): February
Publisher : Sekawan Institut

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.35746/jtim.v8i1.900

The rapid development of e-commerce in Indonesia necessitates recommendation systems that can capture user purchasing patterns accurately, adaptively, and in a data-driven manner. This study implements the FP-Growth algorithm to analyze transaction data from a self-developed essential-goods e-commerce platform. The research dataset consists of 60 user accounts with a total of 600 completed transactions, processed using a Python-based analytical module and au-tomatically integrated into a Laravel backend through a dedicated execution script. The FP-Growth algorithm is applied to generate frequent itemsets and association rules using a min-imum support of 0.01, a minimum confidence of 0.1, and a minimum lift of 1.0. The results indi-cate that the most dominant associative patterns occur among kitchen staple products such as in-stant noodles, chicken eggs, and wheat flour, as well as household cleaning products such as de-tergents and fabric softeners. Several rules exhibit confidence values as high as 0.9615 and lift values up to 4.451, indicating strong and statistically significant relationships between products. System performance evaluation using a Top-4 recommendation scheme shows a Hit Rate of 54.35% and a Recall of 54.35%, demonstrating that the system is able to provide relevant recom-mendations for the majority of transactions. This implementation is shown to improve recom-mendation accuracy while strengthening personalization and cross-selling strategies on essen-tial-goods e-commerce platforms. These findings confirm that FP-Growth is an effective and effi-cient method for identifying empirical purchasing patterns and supporting the development of recommendation systems in small- to medium-scale e-commerce environments.

Title

Found 6 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 6 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 6 Documents
Search