Claim Missing Document
Check
Articles

Found 4 Documents
Search

Analisis Spasial Klaster Atmosfer di Sumatera Utara Menggunakan K-Means Jhon Gabriel Simarmata; Natasha Nainggolan; Elga Sari Tanjung
Jurnal Komputer Teknologi Informasi Sistem Komputer (JUKTISI) Vol. 5 No. 1 (2026): Juni 2026
Publisher : LKP KARYA PRIMA KURSUS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.62712/juktisi.v5i1.936

Abstract

Variabilitas atmosfer jangka pendek di Sumatera Utara pada periode puncak monsun belum dipetakan secara kuantitatif dan spasial. Penelitian ini bertujuan mengidentifikasi dan mengklasifikasikan pola atmosfer pada 28 Desember 2025–1 Januari 2026 menggunakan metode K-Means clustering sebagai solusi analitis berbasis data. Data yang digunakan berasal dari Modern-Era Retrospective Analysis for Research and Applications Version 2 (MERRA-2) dengan variabel suhu udara permukaan, kelembapan spesifik, tekanan permukaan, dan kecepatan angin. Penelitian menggunakan pendekatan kuantitatif deskriptif dengan tahapan praproses data, normalisasi, optimasi jumlah cluster menggunakan Elbow Method, serta evaluasi menggunakan Silhouette Score. Jumlah cluster optimal adalah tiga dengan nilai Silhouette Score global sebesar 0,4089 yang menunjukkan struktur cluster moderat. Hasil penelitian mengidentifikasi tiga pola atmosfer dominan, yaitu Panas Tropis & Sangat Lembap (43,44%), Hangat & Sangat Lembap (36,93%), dan Sejuk & Lembap Sedang (19,63%). Secara spasial, wilayah dataran tinggi berkorelasi dengan kondisi lebih sejuk, sedangkan wilayah pesisir dan kepulauan didominasi kondisi panas dan sangat lembap. Temuan ini menunjukkan bahwa metode K-Means efektif dalam memetakan variabilitas atmosfer spasial jangka pendek secara kuantitatif.
Implementasi Algoritma Greedy untuk Optimasi Protein pada 0/1 Knapsack dalam Diet Defisit Kalori Angelina S. Saragih; Elga Sari Tanjung; Muhammad Rois Lukman Damanik; Adidtya Perdana
Jurnal Riset Informatika dan Inovasi Vol 3 No 12 (2026): JRIIN : Jurnal Riset Informatika dan Inovasi (INPRESS)
Publisher : shofanah Media Berkah

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

Abstract

Program diet defisit kalori memerlukan perencanaan nutrisi yang cermat, khususnya dalam memaksimalkan asupan protein tanpa melebihi batas kalori harian. Permasalahan ini dapat dimodelkan sebagai 0/1 Knapsack Problem di mana kalori bertindak sebagai bobot dan kandungan protein sebagai nilai. Penelitian ini mengimplementasikan dan mengevaluasi tiga strategi Algoritma Greedy, yaitu Greedy by Density (rasio protein/kalori), Greedy by Profit (protein terbesar), dan Greedy by Weight (kalori terkecil) pada dataset 50 menu makanan berbasis Tabel Komposisi Pangan Indonesia (TKPI) dengan kapasitas 1500 kkal. Hasil eksperimen menunjukkan bahwa Greedy by Density menghasilkan total protein tertinggi sebesar 221,2 gram dengan pemanfaatan kapasitas penuh (0 kkal sisa), diikuti Greedy by Profit dengan 176,9 gram, dan Greedy by Weight dengan 128,4 gram. Kompleksitas algoritma secara keseluruhan adalah O(n log n). Penelitian ini membuktikan bahwa strategi berbasis rasio nilai-terhadap-bobot merupakan pendekatan paling efektif dalam optimasi menu diet defisit kalori.
Rancang Bangun Aplikasi Web Identifikasi Teks Akademik dan Non-Akademik Berbasis IndoBERT sebagai Implementasi Pendekatan Natural Language Processing Muhammad Tsabat Muhyiyuddin; Muhammad Zikry Sahendra Chan; Elga Sari Tanjung; Bicanro Gebriyan Panjaitan; Anggia Puteri
Teknik: Jurnal Ilmu Teknik dan Informatika Vol. 6 No. 2 (2026): Oktober: Teknik: Jurnal Ilmu Teknik dan Informatika
Publisher : LPPM Sekolah Tinggi Ilmu Ekonomi - Studi Ekonomi Modern

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/teknik.v6i2.1255

Abstract

The ability to automatically distinguish academic from non-academic texts is an important need in higher education environments. This study aims to design and build a web application that implements the IndoBERT model to classify Indonesian texts into academic or non-academic categories with interpretable output accompanied by classification reasoning. The system is built end-to-end using the PyTorch and HuggingFace Transformers stack for modeling, FastAPI as the backend, and vanilla HTML/CSS/JavaScript for the user interface. The dataset was manually compiled with 209 samples (108 academic and 101 non-academic texts). Model fine-tuning was performed on Google Colab using a Tesla T4 GPU with a learning rate of 2×10⁻⁵, batch size of 16, and eval_loss-based early stopping. The system is also equipped with a zero-shot fallback module so the application remains operational without training. On the internal test set (21 samples), the model achieved 100% accuracy with an F1-score of 1.0000 and an inference speed of 404 samples/second. This research is positioned as a proof-of-concept emphasizing the completeness of the transformer-based NLP pipeline and the interpretability of classification results, rather than a claim of state-of-the-art performance. Discussion of limitations and directions for further development are also presented.
Penerapan DFA untuk Deteksi Serangan TCP SYN dan Ping Flood dengan Visualisasi pada Lingkungan GNS3 Yohanes Gerardus Haga Zai; Zulfahmi Indra; Sevta Triwana Simamora; Elga Sari Tanjung
Teknik: Jurnal Ilmu Teknik dan Informatika Vol. 6 No. 2 (2026): Oktober: Teknik: Jurnal Ilmu Teknik dan Informatika
Publisher : LPPM Sekolah Tinggi Ilmu Ekonomi - Studi Ekonomi Modern

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/teknik.v6i2.1271

Abstract

The rapid development of computer network technology has increased communication efficiency but also raised the risk of cyberattacks, particularly Denial of Service (DoS) attacks such as TCP SYN Flood and ICMP Flood. Conventional Intrusion Detection Systems (IDS) generally rely on static log analysis and lack responsive real-time monitoring visualization. This research aims to design and implement an Intrusion Detection System based on Deterministic Finite Automaton (DFA) for detecting TCP SYN Flood and ICMP Flood attacks in a virtualized network environment. The research method used is an experimental approach through network simulation using Graphical Network Simulator 3 (GNS3). The system utilizes MikroTik Router with TaZmen Sniffer Protocol (TZSP) for packet mirroring, Python with Scapy library for packet analysis and DFA state transition processing, Supabase as a real-time cloud database, and ReactJS as a monitoring dashboard interface. The DFA model classifies network conditions into three states, namely normal, alert, and intrusion, based on the number of packets received per second. The experimental results show that the proposed IDS successfully detects abnormal traffic generated by TCP SYN Flood and ICMP Flood attacks and performs state transitions accurately in real-time. In addition, the integration of Supabase and ReactJS enables the monitoring dashboard to display network security status dynamically without requiring manual page refresh. The implementation of a decoupled architecture also improves system flexibility, responsiveness, and scalability. The findings indicate that the proposed system can provide a lightweight and interactive real-time intrusion monitoring solution for computer network security environments.