Claim Missing Document
Check
Articles

Found 13 Documents
Search

Identifikasi Penyakit Daun Mangga Menggunakan Teknik Pengolahan Citra dan Convolutional Neural Network Rizki Sanria Prasetyo; Agyztia Premana; Nur Ariesanto Ramdhan
Jurnal Publikasi Teknik Informatika Vol. 5 No. 1 (2026): Januari: Jurnal Publikasi Teknik Informatika
Publisher : Lembaga Pengembangan Kinerja Dosen

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/jupti.v5i1.6693

Abstract

The detection of diseases in mango leaves is a significant challenge in Indonesia’s agricultural sector, as it directly affects both the quality and quantity of crop yields. This study aims to develop an image-based classification system for mango leaf diseases using Convolutional Neural Network (CNN) architectures, namely VGG16 and Xception. The dataset used in this research consists of two different datasets. The first dataset includes two classes, healthy and diseased leaves, while the second dataset comprises three classes: Jelangga Fungus, Chlorosis, and Healthy. Data augmentation techniques and the Adam optimizer were applied to enhance model performance. Model evaluation was conducted using a confusion matrix along with precision, recall, and F1-score metrics. The experimental results indicate that VGG16 consistently achieves the best performance, with an accuracy of up to 100% in the two-class classification scenario and 99% in the three-class classification scenario. These findings demonstrate that CNN architectures, particularly VGG16, are effective and reliable for classifying mango leaf diseases based on digital images.
Analisis Keamanan Pertukaran Data Arsitektur Client-Server Menggunakan Protokol JSON Web Token dan Enkripsi AES Mohammad Romdlon; Bambang Irawan; Nur Ariesanto Ramdhan
Smart Comp :Jurnalnya Orang Pintar Komputer Vol 15, No 3 (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.v15i3.10405

Abstract

Pertukaran data pada arsitektur client-server menggunakan protokol HTTP standar sangat rentan terhadap serangan packet sniffing, di mana pihak tidak bertanggung jawab dapat menyadap dan membaca data sensitif secara langsung. Penelitian ini bertujuan untuk meningkatkan keamanan komunikasi data dengan mengimplementasikan metode autentikasi JSON Web Token (JWT) yang dikombinasikan dengan algoritma kriptografi Advanced Encryption Standard (AES). Metode penelitian dilakukan secara eksperimental di laboratorium dengan menguji fungsionalitas aplikasi, menguji ketahanan jaringan dari serangan sniffing menggunakan Wireshark, serta mengukur lonjakan waktu komputasi (overhead response time) menggunakan Postman. Hasil penelitian menunjukkan bahwa sistem berhasil mengamankan jalur komunikasi; data yang disadap melalui jaringan terbukti telah berubah menjadi ciphertext yang tidak dapat dibaca. Selain itu, penambahan lapisan keamanan kriptografi ini terbukti sangat efisien, di mana rata-rata lonjakan waktu respons yang dihasilkan hanya sebesar 4,87 milidetik (dari 9,97 ms menjadi 14,83 ms). Kesimpulannya, kombinasi metode JWT dan enkripsi AES sangat efektif dan direkomendasikan untuk mengamankan pertukaran data pada sistem client-server tanpa mengorbankan performa jaringan secara signifikan.Kata Kunci – Client-Server; Enkripsi AES; JSON Web Token; Keamanan Jaringan; Sniffing 
SISTEM REKOMENDASI TOPIK TUGAS AKHIR MAHASISWA S1 INFORMATIKA BERBASIS CONTENT-BASED FILTERING: STUDI SIMULASI PERBANDINGAN TF-IDF, BM25, DAN LATENT SEMANTIC ANALYSIS Nur Ariesanto Ramdhan
TECHNOVATAR Jurnal Teknologi, Industri, dan Informasi Vol 4 No 3 (2026): Juli
Publisher : Awatara Publisher

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

Abstract

Choosing a final project topic is an early decision that shapes a student's research direction. This study builds and evaluates a prototype recommender for undergraduate Informatics final project topics based on content-based filtering, using each student's interests and course track record as the query. Four methods are compared, namely TF-IDF with cosine similarity, BM25, Latent Semantic Analysis (LSA), and a hybrid of BM25 and LSA, together with a random baseline. The offline evaluation uses 20 synthetic corpora, each with 400 topics from eight Informatics areas and 150 student profiles, and reports Precision@k, MRR, and NDCG@10. Because real data were unavailable and Sentence-BERT could not be downloaded in the research environment, this is a simulation study and does not evaluate Sentence-BERT; its evaluation script is provided. BM25 obtained the highest Precision@5 (0.927), followed by the hybrid (0.914), TF-IDF (0.895), LSA (0.876), and random (0.209); all pairwise differences in Precision@5 remained significant after Holm correction. The hypothesis that latent semantic methods are more robust to term variation was not supported, because LSA declined the most as term variation increased. These findings are preliminary and bounded by the assumptions of the data generator; validation on real final project repositories with expert relevance judgments is still required.