Article Per Year (5 Year)
p-Index From 2021 - 2026
1.203
P-Index
This Author published in this journals
All Journal Insyst : Journal of Intelligent System and Computation Journal of Technology and Informatics (JoTI) Jurnal Atma Inovasia Prosiding Seminar Nasional Universitas Ma Chung Jurnal Teknik Ilmu dan Aplikasi
Windra Swastika
Universitas Ma Chung
Humanities Automotive Engineering Chemical Engineering, Chemistry & Bioengineering Civil Engineering, Building, Construction & Architecture Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Electrical & Electronics Engineering Energy Engineering Environmental Science Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Mechanical Engineering Public Health Social Sciences Other

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Deteksi Objek pada Film Menggunakan Yolo Object Detector dan K-Nearest Neighbor Windra Swastika; Marcellino Agustinus Sinaga
Prosiding Seminar Nasional Universitas Ma Chung Vol. 1 (2021): Prosiding Seminar Nasional Universitas Ma Chung
Publisher : Ma Chung Press

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33479/snumc.v1i.224

Abstract

Peran film sebagai industri budaya merupakan salah satu faktor yang dianggap penting bagi Indonesia. Film harus benar–benar diperhatikan dan dilindungi agar tidak berbalik menjadi pengaruh negatif yang tidak sesuai dan mengakibatkan kemunduran bagi negara Indonesia. Rating usia dari sebuah film merupakan solusi untuk masalah tersebut. Rating usia menunjukkan sentimen pada film tersebut negatif atau positif terutama untuk anak-anak. Pada penelitian ini akan dibuat sebuah sistem untuk mendeteksi objek pada film menggunakan You Only Look Once (YOLO). YOLO digunakan untuk mendapatkan data dari objek yang terdeteksi pada film yaitu untuk objek pistol, pisau, dan rokok dengan jumlah iterasi 50.000. Hasil dari YOLO Mean Average Precision (mAP) adalah 67,14%.
RANCANG BANGUN APLIKASI MONITORING JANTUNG UNTUK KONDISI ARITMIA BERBASIS ANDROID Paulus Lucky Tirma Irawan; Bryan Asa Kristian; Windra Swastika
Jurnal Teknik Ilmu dan Aplikasi Vol. 3 No. 2 (2022): Jurnal Teknik Ilmu dan Aplikasi
Publisher : Politeknik Negeri Malang

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

Abstract

Hingga saat ini penyakit jantung adalah salah satu penyakit yang sangat berbahya. Hal ini dikarenakan kejadiannya sangat sulit untuk diprediksi. Meski begitu terdapat beberapa variabel yang dapat dijadikan acuan untuk mengetahui kapan seseorang dikatakan memiliki potensi yang tinggi untuk mengalami serangan jantung. penyakit jantung bisa dideteksi lebih awal dengan mengetahui gangguan irama jantung (aritmia) yang terjadi. Aritmia merupakan kelainan elektrofisiologi jantung yang dapat disebabkan oleh gangguan sistem konduksi jantung serta gangguan pembentukan dan penghantar impuls. Contoh dari aritmia ini adalah Atrial fibrilasi. Atrial fibrilasi terjadi karena sinyal-sinyal listrik tidak terorganisir dalam atrium dan ventrikel yang menyebabkan detak jantung sangat cepat, lambat dan tidak teratur. Di dalam dunia medis sendiri terdapat teorema waktu kritis yang lebih dikenal dengan Golden Period. Istilah tersebut menandakan waktu kritis maksimal penanganan pasien sebelum terjadi kerusakan permanen dan kematian. Berdasarkan teori ini, kemungkinan penderita selamat dari serangan jantung mendekati nol bila ditangani setelah 10 menit sejak serangan jantung pertama terjadi. Dalam penelitian ini akan dikembangkan sebuah sistem pemantauan berkelanjutan terhadap tanda vital tubuh yang terintegrasi dengan perangkat komunikasi cerdas. Sistem ini dapat menjadi alat pengawas detak jantung secara otomatis dan realtime, sehingga mampu menghindarkan penggunanya dari resiko serangan jantung, dan meningkatkan kewaspadaan akan kondisi jantung. Aplikasi ini akan menggunakan smartband sebagai pembaca detak jantung, serta menggunakan teknologi Location Based Service yang berfungsi sebagai penentu lokasi. Berdasarkan pengujian yang sudah dilakukan seluruh fitur utama termasuk fitur darurat yang berfungsi untuk mengirimkan notifikasi kepada whitelist contact apabila pengguna dalam kondisi aritmia telah bekerja sesuai spesifikasi.
Optimizing KNN: Impact of Distance Metrics and SMOTE on Heart Disease Classification Windra Swastika
Intelligent System and Computation Vol 8 No 1 (2026): INSYST: Journal of Intelligent System and Computation
Publisher : Institut Sains dan Teknologi Terpadu Surabaya (d/h Sekolah Tinggi Teknik Surabaya)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52985/insyst.v8i1.481

Abstract

Heart disease remains the leading cause of mortality worldwide, accounting for approximately 32% of all global deaths. The development of accurate and clinically reliable machine learning-based prediction systems is therefore essential for supporting early clinical decision-making. This study proposes a comprehensive optimization framework for the K-Nearest Neighbor (KNN) algorithm applied to heart disease classification using the UCI Cleveland Heart Disease Dataset. While prior work has addressed class imbalance using the Synthetic Minority Over-sampling Technique (SMOTE) and feature scaling via Min-Max Normalization, no study has simultaneously investigated the effect of distance metric selection and systematic K value optimization in the context of preprocessed imbalanced medical data. This paper makes three contributions: (1) a comparative analysis of three distance metrics, Euclidean, Manhattan, and Minkowski (p=3), applied to KNN after preprocessing; (2) systematic optimal-K identification using Grid Search with Stratified 10-Fold Cross-Validation across all metric-scenario combinations; and (3) a structured ablation study across four preprocessing scenarios to quantify the individual and combined contributions of SMOTE and Min-Max Normalization. Experiments were conducted on 297 samples with 13 clinical features. Results show that the best clinically oriented model (Scenario C: SMOTE + Manhattan, K=9) achieves 81.67% accuracy, 82.14% recall, and 80.70% F1-score. The Minkowski metric in the fully combined scenario (D) achieves the highest AUC of 92.47%, with optimal K=21, a markedly different configuration than Euclidean and Manhattan, which converge at K=1. These findings demonstrate that distance metric choice and K optimization interact significantly, offering practical configuration guidelines for KNN in medical classification tasks.
Optimalisasi Nilai Komersial Produk Pertanian Kelompok Tani Melalui Pemasaran Digital Untuk Mendukung Program Patuwen Kopi Paulus Lucky Tirma Irawan; Windra Swastika; Sultan Arif Rahmadianto
Jurnal Atma Inovasia Vol. 5 No. 6 (2025)
Publisher : Lembaga Penelitian dan Pengabdian pada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24002/jai.v5i6.12037

Abstract

The Patuwen Kopi program initiated by Gereja Kristen Jawi Wetan (GKJW) involves four coffee farmer groups in Malang and Jombang that face significant challenges in product marketing. The primary problems include limited marketing access, dependency on middlemen, and lack of knowledge about digital marketing strategies, resulting in declining economic value of coffee products. This study aims to optimize the commercial value of agricultural products through the development of a website-based digital marketing platform to support the sustainability of the Patuwen Kopi program. The research employed Participatory Rural Appraisal (PRA) methodology consisting of four stages: problem identification using 5 Whys technique, community needs identification, determination of alternative solutions and problem-handling priorities, and identification of ideal solutions. This approach ensures active participation of farmer groups in formulating sustainable solutions. The study successfully developed an e-commerce website application with five main features: landing page for digital marketing media, product page for catalog and ordering, article page for community branding, admin page for dynamic content management, and transactional page for sales recording. The website is integrated with WhatsApp to facilitate communication and transactions, equipped with monthly sales reporting system. The developed digital platform successfully meets the needs of farmer groups in terms of digital marketing, community branding, and transaction recording. The application has undergone limited testing and meets community expectations to support the Patuwen Kopi program, with potential for further development according to future business needs.
Web-Based Automatic Code Evaluation System Using Claude AI for Programming Education Paulus Lucky Tirma Irawan; Windra Swastika
Journal of Technology and Informatics (JoTI) Vol. 8 No. 1 (2026): Vol. 8 N. 1 (2026)
Publisher : Universitas Dinamika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37802/joti.v8i1.1305

Abstract

Algorithm and Programming learning face challenges in providing fast and personalized feedback to students. The manual evaluation process conducted by lecturers requires considerable time, hindering students' iterative learning process. This study aims to develop a web submission platform prototype with automatic feedback based on Claude AI to support and enhance the programming learning process. The research method employs a Research and Development (R&D) approach with four stages: needs analysis, system design and planning, platform implementation, and testing and evaluation. The platform was developed using a PHP backend, MySQL database, and Claude AI integration through RESTful web services with a cascading AI evaluation strategy. Evaluation was conducted on 9 students with 39 submissions for three Java assignments with different difficulty levels. Results show the system successfully provides high-quality feedback with an average response time of 2.8 seconds and 100% evaluation success rate. Score distribution shows average improvement from the first assignment (82.3) to the third assignment (87.1), indicating a positive trend in iterative learning. A satisfaction survey of 8 respondents shows the system interface is user-friendly, and AI feedback helps identify syntax and program logic errors. Students made an average of 3.2 attempts per assignment, demonstrating high engagement in the learning process.
Co-Authors Bryan Asa Kristian Marcellino Agustinus Sinaga Paulus Lucky Tirma Irawan Sultan Arif Rahmadianto