cover
Article Per Year (5 Year)
All
Home Page
OAI Link
Editorial Team
Contact
Reviewer
Google Scholar
Contact Name
Dr. Indrastanti R. Widiasari
Contact Email
editor.aiti@adm.uksw.edu
Phone
-
Journal Mail Official
editor.aiti@adm.uksw.edu
Editorial Address
Kantor Fakultas Teknologi Informasi Jl. O. Notohamidjojo 1-10 Salatiga, Jawa Tengah 50711
Location
Kota salatiga,
Jawa tengah
INDONESIA
Aiti: Jurnal Teknologi Informasi
Published by Universitas Kristen Satya Wacana
ISSN : 16938348     EISSN : 26157128     DOI : https://doi.org/10.24246/aiti
Core Subject : Science,
Computer Science & IT
AITI: Jurnal Teknologi Informasi is a peer-review journal focusing on information system and technology issues. AITI invites academics and researchers who do original research in information system and technology, including but not limited to: Cryptography Networking Internet of Things Big Data Data Science Software Engineering Information System Web Programming Mobile Application Service System Artificial Intelligence Digital Image Processing Machine Learning Deep Learning Geographic Information System Context Aware System Management Information System Software-defined Network
Arjuna Subject : Ilmu Komputer (semua kategori) - Ilmu Komputer (Lain-Lain)
Articles 185 Documents
 15   16   17   18   19 
Ekstraksi informasi leksikal bahasa Kerinci dari kamus dwibahasa untuk klasifikasi Part-of-Speech otomatis Agung Kharisma Hidayah; Dandi Sunardi; Sri Handayani
AITI Vol 23 No 2 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i2.334-347

Abstract

Bahasa Kerinci merupakan salah satu bahasa daerah di Indonesia yang tergolong low-resource language, sehingga ketersediaan sumber daya linguistik digital sangat terbatas. Penelitian ini bertujuan untuk mengekstraksi informasi leksikal dari kamus dwibahasa Indonesia–Kerinci dan melakukan klasifikasi kelas kata (Part-of-Speech/POS) secara otomatis menggunakan pendekatan berbasis aturan (rule-based). Proses penelitian dimulai dengan mengonversi kamus dwibahasa Indonesia–Kerinci ke dalam format dataset terstruktur yang memuat lema, padanan kata, serta informasi linguistik terkait. Selanjutnya, dilakukan klasifikasi kelas kata secara otomatis menggunakan pendekatan rule-based dengan memanfaatkan padanan bahasa Indonesia sebagai acuan untuk menentukan POS pada lema bahasa Kerinci. Hasil penelitian menghasilkan dataset digital terstruktur beserta label POS otomatis yang dapat dijadikan sumber daya awal untuk pemrosesan bahasa alami (Natural Language Processing/NLP) dalam bahasa Kerinci. Kontribusi utama penelitian ini adalah menyediakan sumber daya linguistik digital awal untuk bahasa daerah yang masih minim data, serta menawarkan strategi klasifikasi POS berbasis leksikal yang sederhana dan dapat direplikasi pada bahasa daerah lain.
Rancang bangun sistem FlowTrackr: Aplikasi manajemen test case dan bug report berbasis web Desril Fatra Aditama; Imelda Imelda
AITI Vol 23 No 2 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i2.290-304

Abstract

Software testing is an essential stage to ensure application quality before deployment. Manual testing practices by Quality Assurance (QA) teams often face challenges, such as scattered test results and bug reports across multiple media, making them difficult to trace and prone to miscommunication. This study designs FlowTrackr, a web-based integrated application for centralized test case management and bug reporting. The development followed the Waterfall methodology, covering requirements analysis, system design, implementation, and testing. Functional testing across 10 scenarios confirmed that all features operated according to specifications, while User Acceptance Testing (UAT) involving 5 QA testers demonstrated that the system was easy to use and more efficient than manual methods. Compared with tools such as Jira, TestRail, and Bugzilla, which tend to be more complex, FlowTrackr offers a lightweight, practical solution for small teams. Therefore, FlowTrackr can serve as an alternative to support manual QA processes, making them more structured and well-documented.
Lung cancer prognosis based on salivary biomarkers using Graph Convolutional Networks Gilang Raka Rayuda Dewa; Raisa Imani Sani; Ady Syamsuri; Charles Agustin; Muhammad Agni Catur Bhakti; Ariana Tulus Purnomo
AITI Vol 23 No 2 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i2.245-260

Abstract

Lung cancer remains the leading factors that incur cancer-related deaths worldwide, mainly due to late-stage detection. In 2022, lung cancer affected almost 2.5 million people, with mortality of more than 1.8 million. However, existing prognostic methods are typically invasive, costly, and time-consuming, hindering effective early detection. Therefore, this research proposes a non-invasive prognostic approach using salivary biomarkers to detect lung cancer via Graph Convolutional Networks (GCNs). By transforming features into graph node representations, the proposed algorithm can model feature dependencies and topological relationships, enabling more effective pattern recognition than conventional classifiers. The proposed algorithm also applies feature selection to reduce computational complexity. The evaluation results show that the proposed algorithm achieves 95.65% accuracy, a macro F1-score of 95.62%, and a Matthews Correlation Coefficient of 0.9434. A comparative analysis shows that the proposed algorithm outperforms other graph-based architectures in terms of classification performance and computational complexity.
Implementasi YOLOv11 untuk pengenalan nominal uang Rupiah dengan konversi suara otomatis Muhamad Alvin Andriyanto; Yani Parti Astuti
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.348-362

Abstract

Penelitian ini bertujuan mengembangkan sistem pendeteksi nominal uang rupiah berbasis computer vision yang dapat memberikan keluaran suara otomatis sebagai dukungan aksesibilitas bagi penyandang tunanetra. Model dikembangkan menggunakan algoritma YOLOv11 yang disesuaikan untuk mengenali tujuh kelas nominal uang kertas emisi 2022. Dataset publik digunakan sebagai data latih, validasi, dan uji, kemudian diproses melalui transfromasi dan augmentasi untuk meningkatkan generalisasi model. Pelatihan dilakukan menggunakan konfigurasi terkontrol dengan optimisasi AdamW dan strategi pencegahan overfitting. Evaluasi kinerja dilakukan menggunakan metrik akurasi, precision, recall, F1-score, mean Average Precision, confusion matrix, serta ROC-AUC. Hasil penelitian menunjukkan bahwa model mencapai performa sangat tinggi dengan akurasi 99,64% dan mAP 0,9935, menandakan kemampuan identifikasi nominal uang yang konsisten pada seluruh kelas. Implementasi sederhana berbasis OpenCV dan konversi suara menggunakan gTTS membuktikan bahwa model dapat beroperasi secara real-time dan memberikan umpan balik audio secara langsung. Temuan ini menunjukkan bahwa YOLOv11 efektif untuk pengenalan uang rupiah dan berpotensi dikembangkan lebih lanjut pada aplikasi aksesibilitas bagi penyandang tunanetra.
Deteksi URL phishing menggunakan kombinasi model XGBoost dan Random Forest Ilham Maulana Hadinanda; Sindhu Rakasiwi
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.363-377

Abstract

Phishing is one form of cyber attack in today's digital era. This attack uses manipulation techniques to deceive users, so that they unknowingly disclose sensitive data through fake websites that mimic the appearance of official websites. Detecting phishing URLs is an important step in improving cyber security and reducing user losses. This study uses an ensemble learning approach by combining two models, Extreme Gradient Boosting (XGBoost) and Random Forest, into one combined model. Both models were trained using a dataset consisting of 11,430 URLs, divided equally between phishing and legitimate classes. The model was tested using accuracy, precision, recall, and F1-score metrics. The results of the experiment show that the combined XGBoost and Random Forest model produced the highest accuracy, at 0.944. This is superior to the single XGBoost (0.942) and Random Forest (0.928) models. These findings reinforce the evidence that the ensemble approach provides better generalization than single models for phishing detection.
Sistem deteksi hama dan penyakit daun kelapa dengan metode CNN menggunakan arsitektur MobileNetV2 Olivia Kanakang; Audy Kenap; Medi Tinambunan
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.378-391

Abstract

Tanaman kelapa merupakan sumber pendapatan penting bagi masyarakat di Kecamatan Melonguane Timur, namun rentan terhadap serangan hama dan penyakit daun yang dapat mengganggu proses fotosintesis serta menurunkan hasil panen. Petani seringkali terlambat dalam mendeteksi gejala awal sehingga penanganan menjadi kurang optimal. Meskipun penelitian sebelumnya yang menggunakan Convolutional Neural Network (CNN) menunjukkan tingkat akurasi yang baik, masih sedikit penelitian yang mengimplementasikan sistem berbasis web khusus untuk daun kelapa di wilayah terpencil, serta belum banyak yang melaporkan detail hiperparameter secara lengkap untuk mendukung reproduktibilitas penelitian. Oleh karena itu, penelitian ini bertujuan untuk mengembangkan aplikasi berbasis situs web dalam mendeteksi hama dan penyakit daun kelapa menggunakan CNN dengan arsitektur MobileNetV2. Dataset yang digunakan terdiri dari 1.000 citra dengan empat kategori, yaitu ulat api, daun kering, daun layu, dan daun menguning. Model dilatih menggunakan metode transfer learning dengan 25 epoch, ukuran batch 16, laju pembelajaran 0,0001, optimizer Adam, dan fungsi kerugian categorical cross-entropy. Hasil pengujian menunjukkan bahwa model mencapai akurasi sebesar 96% (193 dari 200 data uji benar), dengan precision 96%, recall 96%, dan F1-score 96%, sehingga efektif untuk klasifikasi yang cepat dan akurat dalam penerapan nyata.
Optimizing CNN for burn severity classification using hyperband Vallerie Britania Sepang; Winsy Christo Deilan Weku; Aditya Lapu Kalua
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.407-419

Abstract

Burn injuries require accurate severity diagnosis to determine appropriate treatment and reduce mortality. However, conventional clinical assessment is susceptible to subjectivity and diagnostic errors. This study aims to optimize a Convolutional Neural Network (CNN) model for classifying burn severity using the Hyperband hyperparameter optimization algorithm. A dataset of 1,040 burn images, collected from primary and secondary sources, was categorized as first-, second-, or third-degree burns. A MobileNet-based CNN model was trained using both default and Hyperband-optimized hyperparameters and evaluated using 10-fold cross-validation. The results demonstrate that hyperparameter optimization substantially improved classification performance. The optimized model achieved an accuracy of 93.57%, compared with 82.14% for the baseline model. Improvements were also observed in precision, recall, and F1-score, indicating better classification performance and reduced prediction errors. These findings demonstrate that Hyperband can effectively optimize CNN hyperparameters for burn severity classification. The proposed approach has the potential to support more consistent and accurate burn assessment. Future research should investigate advanced optimization strategies and larger, more diverse datasets to improve model generalization and robustness further.
Integrasi Content Provider untuk interoperabilitas aplikasi Java Android dan Kivy Python Hosea Nicolas A. N. P. Putra; Ariya Dwika Cahyono
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.392-406

Abstract

This study aims to enable data interoperability between Java-based Android applications and Kivy-based Python applications by implementing the Content Provider mechanism. The research addresses the challenge of exchanging data between applications developed using different programming languages and frameworks, particularly the complexity of establishing direct communication between Java-based Android and Kivy applications. The proposed approach employs Android Content Providers as an intermediary mechanism to provide controlled access to application data and facilitate data exchange between heterogeneous applications. The implementation was evaluated based on data exchange functionality, communication latency, and data consistency. The results demonstrate that integrating Content Providers successfully enables data exchange between Java-based Android and Kivy applications while maintaining data consistency. The measured communication latency was within an acceptable range, indicating that the proposed mechanism can support inter-application communication without significant performance overhead. These findings demonstrate that Content Providers can improve interoperability between heterogeneous application platforms and support hybrid application development. Therefore, the proposed approach provides a practical solution for overcoming data communication barriers between Java-based Android and Kivy applications while increasing flexibility in mobile application development.
Penerapan metode SVM dengan fitur HSV dan HOG pada klasifikasi jenis tanaman hias Aglaonema Leni Kusneti; Sri Andayani
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.482-495

Abstract

Aglaonema merupakan salah satu tanaman hias populer di Indonesia yang memiliki keanekaragaman bentuk, warna, dan corak daun. Variasi visual yang mirip antar varietas seringkali menyulitkan proses identifikasi secara manual. Penelitian ini bertujuan untuk mengklasifikasikan jenis Aglaonema menggunakan metode Support Vector Machine (SVM) dengan fitur segmentasi warna HSV dan fitur ekstraksi bentuk HOG. Dataset terdiri dari 1.400 citra daun yang terbagi ke dalam tujuh jenis Aglaonema, diantaranya Big Roy, Cochin, Dud Anjamani, Heng Heng, Snow White, Lipstik Merah dan Butterfly. Hasil pengujian metode SVM kernel linear pada validasi silang 5-K-Fold Cross Validation menunjukkan accuracy tertinggi pada Dud Anjamani sebesar 97,21% dan terendah pada Cochin sebesar 92,64%. Rata-rata precision tertinggi dicapai oleh Butterfly sebesar 95,79%, dan terendah pada Cochin sebesar 67,99%, sementara itu rata-rata recall tertinggi pada Big Roy sebesar 91,94%, dan terendah pada Butterfly dan Lipstik Merah sebesar 79,00%. Secara keseluruhan, model mencapai rata-rata accuracy 95,47%, precision 85,51%, dan recall 84,21%.
Analisis emosi pengguna untuk optimasi desain antarmuka Admisi UKSW dengan pendekatan Kansei Engineering Danang Tri Widiatmoko; Hendry Hendry; Krismiyati Krismiyati
AITI Vol 23 No 3 (2026)
Publisher : Fakultas Teknologi Informasi Universitas Kristen Satya Wacana

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24246/aiti.v23i3.511-532

Abstract

Penelitian ini merespons kebutuhan pembaruan antarmuka Admisi Online UKSW, sistem pendaftaran daring yang tidak diperbarui sejak 2016 dan dinilai tidak sesuai ekspektasi visual Generasi Z. Tujuan utamanya adalah merancang rekomendasi UI berbasis preferensi emosional pengguna muda menggunakan pendekatan Kansei Engineering (KE) Tipe I. Persepsi pengguna digali melalui pertanyaan terbuka dan dianalisis menggunakan TF-IDF untuk mengekstraksi Kansei words, yang kemudian dinilai melalui skala Semantic Differential dan dianalisis menggunakan Principal Component Analysis (PCA), Coefficient Correlation Analysis (CCA), Factor Analysis (FA), dan PLS Regression. Hasil menunjukkan bahwa kata seperti “modern”, “menarik”, “mudah”, “simple”, dan “kekinian” berkorelasi kuat dengan elemen visual seperti warna kontras, ikon, ukuran huruf, dan bentuk tombol. Purwarupa UI dan matriks desain disusun berdasarkan hasil analisis dan divalidasi oleh pakar melalui expert evaluation. Penelitian ini menghasilkan kerangka kerja generik penerapan KE Tipe I yang dapat direplikasi, serta membuktikan efektivitas pendekatan KE dalam merancang UI sistem digital berbasis emosi di institusi pendidikan tinggi.

Page 18 of 19 | Total Record : 185

 15   16   17   18   19 

Filter by Year

2016 2026


Reset
Filter By Issues
All Issue Vol 23 No 3 (2026) Vol 23 No 2 (2026) Vol 23 No 1 (2026) Vol 22 No 2 (2025) Vol 22 No 1 (2025) Vol 21 No 2 (2024) Vol 21 No 1 (2024) Vol 20 No 2 (2023) Vol 20 No 1 (2023) Vol 19 No 2 (2022) Vol 19 No 1 (2022) Vol 18 No 2 (2021) Vol 18 No 1 (2021) Vol 17 No 2 (2020) Vol 17 No 1 (2020) Vol 16 No 2 (2019) Vol 16 No 1 (2019) Vol 15 No 2 (2018) Vol 15 No 1 (2018) Vol 14 No 2 (2017) Vol 14 No 1 (2017) Vol 13 No 2 (2016) Vol 13 No 1 (2016) More Issue