Claim Missing Document
Check
Articles

Found 7 Documents
Search

The Implementation of Convolutional Neural Network in Recognize Lontara Text Arthanugraha, Wahyu; Zainuddin, Zahir; Arda, Abdul Latief
Jurnal Informatika Vol 11, No 2 (2024): October
Publisher : Universitas Bina Sarana Informatika

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31294/inf.v11i2.20328

Abstract

The introduction of the Lontara text that is applied is still conventional in that it comes from books or teaching materials and there is no application for a translator of the Lontara text, making it difficult for students, especially those from outside the Bugis tribe. This research aims to measure the ability and apply the CNN method in the context of introducing the Lontara text. The input that use in this research is words taken from the original text or writing manuscripts of the ancient literature of the Bugis tribe "Sureq Maqkelluqna Nabittaq", a book that was used as a tool to spread Islam in 1611. The model used is the VGG-16 architecture. This research uses the R&D (Research and Development) method which is used to create products and test their level of effectiveness. As for the research results obtained, the system accuracy obtained in classifying the Lontara text using the VGG16 architecture was 97,65%. In addition, the system is also able to display the translation of the Lontara text according to the database input.
Sistem Informasi Geografis Pemetaan Objek Wisata Di Kabupaten Barru Berbasis Web Ahmad; Ahmad Ahmad; Marlina; Hasnawati; Masnur; Wahyu Artanugraha
Elkom: Jurnal Elektronika dan Komputer Vol. 18 No. 2 (2025): Desember : Jurnal Elektronika dan Komputer
Publisher : STEKOM PRESS

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.51903/elkom.v18i2.3065

Abstract

Barru Regency tourism has a variety of tourist attractions, but information related to the location and potential of tourism has not been digitally integrated, making it difficult for tourists and the government to access data. This study aims to design and build a web-based Geographic Information System (GIS) that can map the location of tourist attractions in Barru Regency interactively, easily accessible, and equipped with supporting information in the form of descriptions, types of tourism, photos, and travel routes. The research method used is Research and Development (R&D) with the stages of tourist attraction surveys, interface design, feature development, and system testing. The results of the study are in the form of a website "Web-Based Geographic Information System for Mapping Tourist Attractions in Barru Regency" which is able to present tourist information systematically and easily understood. The conclusion of this study shows that the developed system can be a supporting medium in disseminating tourism information, helping tourists find tourist locations, and supporting the local government in managing and developing the tourism sector based on spatial data. This application also has the potential to be an educational and promotional tool to increase tourist visits to Barru Regency  
Klasifikasi Tingkat Kematangan Tomat Menggunakan Algoritma Knn (K-Nearest Neighbor) Berbasis Citra Digital Muhammad Suryanto Rustam; Marlina; Mughaffir Yunus; Andi Wafiah; Wahyu Artanugraha
Jurnal Publikasi Ilmu Komputer dan Multimedia Vol. 4 No. 3 (2025): September: Jurnal Publikasi Ilmu Komputer dan Multimedia
Publisher : Pusat Riset dan Inovasi Nasional

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.55606/jupikom.v4i3.5275

Abstract

The problem raised in this study is the difficulty of accurately classifying tomato ripeness levels if only relying on visual observation, so a more objective computational method is needed. This study aims to design and implement a tomato ripeness classification system using the K-Nearest Neighbor (KNN) method based on digital image processing. The dataset used consists of 300 tomato images taken from agricultural land in Enrekang Regency, South Sulawesi, with an even distribution in each ripeness category. The method used includes taking tomato images, resizing the images to 200×200 pixels, extracting RGB and HSV color features, and normalizing pixel values. The features used in the classification are the average values ​​of Hue, Saturation, and Value of each image. The KNN algorithm with parameter K = 3 is applied to compare the Euclidean distance between the test data and the training data. The test results show that the accuracy per category reaches 100%, and the overall accuracy is 90%. These findings prove that the combination of HSV and KNN color models is effective in distinguishing tomato ripeness levels, and has the potential to be implemented in automated sorting systems to improve post-harvest efficiency in the agricultural sector.
Transformasi Akuntabilitas Keuangan Keluarga Bugis: Pengembangan Template Excel Berbasis Nilai Siri’na Pacce Hernianti Harun; Wahyu Arthanugraha; Siti Rahmawati Harun
Owner : Riset dan Jurnal Akuntansi Vol. 10 No. 3 (2026): Periode Juli 2026
Publisher : Politeknik Ganesha Medan

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.33395/owner.v10i3.3315

Abstract

The financial management of traditional Bugis weddings faces serious challenges: the complexity of traditional ceremonies, which involve many parties and funding sources, is not supported by an adequate record-keeping system, resulting in financial transparency, efficiency, and accountability within families often being neglected. This study aims to: (1) develop a user-friendly Excel-based family expense accountability model tailored to the Bugis cultural context; (2) design an integrated Excel template along with usage guidelines for digital recording of traditional weddings; and (3) evaluate the effectiveness of the Excel template’s implementation in enhancing transparency, efficiency, and accountability in family financial management. The study employed a mixed-methods approach with a sequential explanatory design, involving 60 respondents in the quantitative phase and 8 key informants in the qualitative phase, supported by two sessions of focus group discussions (FGDs) and participatory observation. Instrument validation showed that all items were valid (r = 0.766–0.903) and reliable (? = 0.692–0.872). Multiple regression results yielded a coefficient of determination R² = 0.710 (F = 26.462; p < 0.001), with perceived ease of use (PEOU) as the most dominant predictor (? = 0.283; p = 0.004). The study produced an eight-sheet Excel template designed based on Bugis traditional rituals, with an increase in the average accountability score
Accuracy and Stability Trade-off in RSSI Window Aggregation for MLP-Based Wi-Fi Indoor Positioning Wahyu Arthanugraha; Abdul Halim Ismail; Shahrul Nizam Yaakob
Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol 14, No 3: September 2026
Publisher : IAES Indonesian Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52549/ijeei.v14i3.8256

Abstract

Wi-Fi RSSI fingerprinting is practical for indoor positioning, but short-term signal fluctuations can reduce classification consistency. This study evaluates how fixed RSSI window size affects the accuracy, stability, and sample availability of an MLP-based indoor positioning system. RSSI from three Access Points was collected at 45 ordered corridor positions. Non-overlapping windows with n=1,3,5,7,9,11,15,21 and 29 were represented by the mean and standard deviation of three RSSI channels. The MLP was assessed using stratified 5-fold cross-validation with supplementary order-blocked validation. Relative to the n=1 baseline Accuracy of 68.24%, n=7 achieved 90.60 ± 0.71% Accuracy, 95.08 ± 1.31% Top-2 Accuracy, 0.9047 ± 0.0060 Macro-F1, and 0.4319 ± 0.1364 Ordinal Error. Larger windows achieved higher mean Accuracy but sharply reduced window-level samples and increased fold-to-fold variability. Order-blocked validation produced more conservative estimates and shifted the strongest performance region toward n=9 to n=11. These findings show that fixed-window selection should consider predictive performance, evaluation stability, sample availability, and validation design rather than maximum Accuracy alone.
Pengenalan Pola Batik dan Tenun Sutera Bugis Menggunakan Convolutional Neural Network (CNN): Recognition of Bugis Batik and Silk Weaving Patterns Using a Convolutional Neural Network (CNN) Pawelloi, Andi Irmayani; Arifuddin, Arifuddin; Arthanugraha, Wahyu; Selao, Ahmad; Wafiah, Andi
MALCOM: Indonesian Journal of Machine Learning and Computer Science Vol. 6 No. 4 (2026): MALCOM October 2026
Publisher : Institut Riset dan Publikasi Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.57152/malcom.v6i4.3058

Abstract

Kabupaten Wajo memiliki warisan tenun sutra Bugis yang memiliki nilai filosofis tinggi. Identifikasi motif kain ini masih dilakukan secara manual. Cara tradisional tersebut memicu kesalahan klasifikasi dan menghambat digitalisasi budaya. Riset ini membangun sistem pengenalan motif berbasis Convolutional Neural Network (CNN) sekaligus mengukur tingkat akurasinya. Peneliti menerapkan metode Research and Development (R&D) melalui pendekatan prototipe. Peneliti mengumpulkan 3.508 citra dari observasi, katalog, dan media sosial. Himpunan data ini mencakup tujuh kelas motif dan satu kelas Non-Motif. Peneliti membagi data menjadi 70% untuk pelatihan, 15% untuk validasi, dan 15% untuk pengujian. Setelah prapemrosesan, model menjalani pelatihan transfer learning menggunakan arsitektur modifikasi MobileNetV2. Pelatihan berlangsung selama 50 epoch dengan input 224x224x3, batch size 32, dan optimizer Adam. Peneliti lalu mengintegrasikan model ke aplikasi web berbasis Flask. Kontribusi utama riset ini menghadirkan sistem identifikasi otomatis multi-kelas untuk mendukung digitalisasi tekstil Bugis. Pengujian terhadap 533 citra mencatat akurasi 97,76 persen serta skor makro precision, recall, dan F1-score sebesar 0,98. Kelas Bukan Motif, Bunga Lagosi, dan Sobbi KDI meraih F1-score sempurna sebesar 1,00. Temuan ini membuktikan bahwa sistem mampu mengenali motif secara presisi dan efektif melestarikan budaya Bugis
Accuracy and Stability Trade-off in RSSI Window Aggregation for MLP-Based Wi-Fi Indoor Positioning Wahyu Arthanugraha; Abdul Halim Ismail; Shahrul Nizam Yaakob
Indonesian Journal of Electrical Engineering and Informatics (IJEEI) Vol. 14 No. 3: September 2026
Publisher : IAES Indonesian Section

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.52549/ijeei.v14i3.8256

Abstract

Wi-Fi RSSI fingerprinting is practical for indoor positioning, but short-term signal fluctuations can reduce classification consistency. This study evaluates how fixed RSSI window size affects the accuracy, stability, and sample availability of an MLP-based indoor positioning system. RSSI from three Access Points was collected at 45 ordered corridor positions. Non-overlapping windows with n=1,3,5,7,9,11,15,21 and 29 were represented by the mean and standard deviation of three RSSI channels. The MLP was assessed using stratified 5-fold cross-validation with supplementary order-blocked validation. Relative to the n=1 baseline Accuracy of 68.24%, n=7 achieved 90.60 ± 0.71% Accuracy, 95.08 ± 1.31% Top-2 Accuracy, 0.9047 ± 0.0060 Macro-F1, and 0.4319 ± 0.1364 Ordinal Error. Larger windows achieved higher mean Accuracy but sharply reduced window-level samples and increased fold-to-fold variability. Order-blocked validation produced more conservative estimates and shifted the strongest performance region toward n=9 to n=11. These findings show that fixed-window selection should consider predictive performance, evaluation stability, sample availability, and validation design rather than maximum Accuracy alone.