Garuda - Garba Rujukan Digital

PENERAPAN VIDEO PEMBELAJARAN BAGIAN DAN FUNGSI PADA TANAMAN UNTUK KELAS 4 SD DI SD MUHAMMADIYAH 3 DENPASAR: Implementation on Video Tutorial of Plants Parts and Function for 4th Grade Elementary School Students at Muhammadiyah 3 Elementary School Denpasar Welda Welda; Aniek Suryanti Kusuma; Ayu Manik Dirgayusari; I Nyoman Agus Suarya Putra; I Gusti Agung Indrawan
PUAN INDONESIA Vol. 5 No. 1 (2023): Jurnal Puan Indonesia Vol 5 No 1 Juli 2023
Publisher : ASOSIASI IDEBAHASA KEPRI

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37296/jpi.v5i1.148

Technological advances and the current pandemic situation require SD Muhammadiyah 3 Denpasar to evaluate the current learning system. The current learning system is less effective due to advances in technological developments, as well as the prolonged pandemic situation that has forced the learning system to be carried out online. Students are still confused about participating in online learning, and are often left behind in effective online learning because the gadgets students usually use for online learning are sometimes being used by their parents or other circumstances. Students need a source of literature or other sources of teaching materials that allow students to review the teaching materials after study hours. Therefore, this community service activity is carried out to provide an effective alternative to be applied to students, namely by making animated videos covering natural science subject matter regarding the functions and parts of plants. This community service activity begins with survey and preparation activities, analysis and development of learning videos, and then conducts outreach and training on how to use these learning videos. Based on the results of the training conducted, students and teachers can properly use this learning video as an alternative reference for learning materials.

Analisis Faktor yang Mempengaruhi Customer Loyalty dalam Menggunakan Uang Elektronik Berbasis Server Ni Made Mila Rosa Desmayani; I Gusti Agung Indrawan
Ekuitas: Jurnal Pendidikan Ekonomi Vol. 10 No. 1 (2022)
Publisher : Fakultas Ekonomi Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/ekuitas.v10i1.30306

Electronic money providers can be divided into two. First is a provider of electronic money from banks in the form of a physical chip card. Examples are e-money from Bank Mandiri and Flazz from BCA. The second is a server-based, non-banking e-money provider. Examples of products are GoPay, OVO and DANA. Competition is very tight to dominate the electronic money market share, encouraging server-based electronic money providers to compete to expand services, and provide facilities and discounts to its users. This has the potential to seize competitors' marketshare, where electronic money users will switch to other providers. This article aims to analyze the factors that affect customer loyalty of server-based electronic money users in Denpasar, Bali. The object of this research is the factors that affects customer loyalty in using server-based electronic money. This research uses modified technology acceptance model. The collection of data on the perception of electronic money was carried out using a questionnaire. The results obtained from this questionnaire are: (1) perceived user convenience, (2) perceived usefulness, (3) perceived price suitability, (4) perceived discount has a positive dan significant effect on user loyalty in using server-based e-money.

PKM Optimalisasi Pengelolaan Sampah Melalui Pemilahan Pada Sumber Timbulannya di Desa Pecatu, Kecamatan Kuta Selatan, Kabupaten Badung, Bali Sandika, I Kadek Budi; Ariasih, Ni Kadek; Sutarwiyasa, I Ketut; Lesmana, Putu Surya Wedra; Ginantra, Ni Luh Wiwik Sri Rahayu; Widiartha, Komang Kurniawan; Marlinda, Ni Luh Putu Mery; Indrawan, I Gusti Agung
Jurnal Pengabdian Masyarakat Sains dan Teknologi Vol. 1 No. 4 (2022): Desember : Jurnal Pengabdian Masyarakat Sains dan Teknologi
Publisher : Fakultas Teknik Universitas Cenderawasih

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.58169/jpmsaintek.v1i4.61

Pemerintahan desa Pecatu belum mampu menangani sampah pada unit jasa pengelolaan sampah secara maksimal karena sampah belum terpilah di sumber timbulannya, mesin yang kurang bagus, serta belum menemukan formula tepat menangani masalah residu. Tujuan Kegiatan PKM yaitu melakukan optimalisasi pengelolaan sampah melalui proses pemilihan. proses edukasi kepada masyarakat sebagai target dari transfer knowledge tata cara pemilahan sampah, karena masyarakat merupakan sumber timbulan dari pengelolaan sampah. Kegiatan PKM ini memberikan solusi yang ditawarkan dengan sumber daya yang dimiliki adalah sosialisasi, edukasi dan pendampingan pemilahan sampah di sumber timbulan sampah/pelanggan. Hasil penelitian berfokus pada kegiatan pendampingan program pemilahan sampah ditargetkan pada 5% pelanggan unit jasa pengelolaan sampah, Tahapan pengabdian yang direncanakan adalah persiapan, FGD unsur pemerintahan desa dinas dan desa adat, sosialisasi pemilahan sampah, monitoring pelaksanaan pemilahan sampah disumbernya, serta evaluasi kegiatan. Hasil PKM menunjukkan tingkat pemahaman dan konsistensi masyarakat pada proses pemilahan sampah rerata adalah 63%.

ANALISIS PEMILIHAN CLUSTER OPTIMAL DALAM SEGMENTASI PELANGGAN TOKO RETAIL Murpratiwi, Santi Ika; Agung Indrawan, I Gusti; Aranta, Arik
Jurnal Pendidikan Teknologi dan Kejuruan Vol. 18 No. 2 (2021): Edisi Juli 2021
Publisher : Universitas Pendidikan Ganesha

Saat ini pemanfaatan data menjadi fokus dalam bidang pemasaran khususnya untuk menyusun strategi. Agar strategi pemasaran bisa tepat sasaran dibutuhkan segmentasi pelanggan. Data mining khususnya clustering mampu membantu proses segmentasi pelanggan. Dalam penelitian ini, data mining diimplementasikan untuk segmentasi pelanggan UD. XYZ dengan metode K-Means, K-medoids, dan Means.. Tujuan penelitian ini adalah mencari metode dan nilai k terbaik yang dihasilkan dari tiga metode clustering. Penelitian ini menyajikan proses Data Mining dengan menggabungkan model RFM dengan algoritma clustering K-Medoids, X-Means, dan K-Means. Dataset yang telah diimplementasikan ke dalam model RFM digunakan sebagai bahan pengolahan data. Data transaksi dengan jumlah 153.492 diimplementasikan ke dalam model RFM menjadi 10.145 data untuk dilakukan identifikasi pelanggan potensial. Inisialisasi cluster awal pada metode K-Medoids, X-Means, dan K-Means dilakukan secara random. Nilai k dalam penelitian ini diinisialisasi dari 1 sampai 10. Nilai k diimplementasikan secara berulang dan dihitung validasi cluster menggunakan metode David Bouldin Index (DBI) dan jaraj rata-rata cluster dengan centroid. Hasil penelitian menunjukkan K-medoids memiliki nilai validitas yang lebih baik dibandingkan dengan X-Means dan K-Means. Rata-rata nilai DBI yang dihasilkan metode K-Medoids adalah 0,540778. Jumlah cluster terbaik yang dihasilkan adalah 5 cluster, hal ini ditentukan dengan mempertimbangkan jumlah persebaran data pada k = 5 yang menghasilkan nilai sama pada metode K-Medoids, X-Means, dan K-Means. Tingkatan pelanggan yang terbentuk adalah About To Sleep, Customer Needing Attention, Recent Customer, Potential Loyalist, dan Loyal Customers.

Comparison of Deep Learning Methods for Detecting Tuberculosis Through Chest X-Rays Udayana, I Putu Agus Eka Darma; Indrawan, I Gusti Agung; Prawira, I Made Karang Satria
Journal of Computer Networks, Architecture and High Performance Computing Vol. 6 No. 3 (2024): Articles Research Volume 6 Issue 3, July 2024
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v6i3.4345

Chronic diseases are the leading cause of death worldwide, accounting for 73% of deaths in 2020. Tuberculosis (TB), caused by the bacterium Mycobacterium tuberculosis, is one of these diseases and has a significant impact on countries with a high TB burden due to a lack of radiologists and medical equipment. Early diagnosis of TB is crucial but challenging because of its similarity to lung cancer and the shortage of radiologists. A semi-automatic TB detection system is needed to support medical diagnosis and improve public health services. Deep learning technology, such as Convolutional Neural Networks (CNN), offers an effective solution for disease diagnosis with high accuracy. This study compares deep learning methods using an 8-layer CNN and VGG-19, both enhanced with Histogram Equalization (HE) for improved image quality. The study utilizes chest X-ray images of normal lungs and TB-affected lungs from Kaggle. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Results indicate that the VGG-19 model outperforms the 8-layer CNN across all evaluation metrics, achieving an accuracy of 72.00% compared to 65.00% for the 8-layer CNN. VGG-19 also demonstrates better precision, recall, and F1-score, making it a more suitable choice for TB detection with enhanced image quality.

Classification of Cavendish Banana Ripeness With CNN Method Tjokorda Istri Agung Pandu Yuni Maharani; I Gusti Agung Indrawan; Gede Dana Pramitha; Christina Purnama Yanti; I Made Marthana Yusa
Indonesian Journal of Data and Science Vol. 6 No. 2 (2025): Indonesian Journal of Data and Science
Publisher : yocto brain

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56705/ijodas.v6i2.259

Cavendish bananas are one of the most widely consumed tropical fruits in Indonesia due to their sweet taste and high nutritional content. However, as they ripen, the sugar content in bananas increases, which can be a problem for diabetics. To help diabetics choose bananas with the right level of ripeness, this study developed a Cavendish banana ripeness classification model using artificial intelligence technology, namely the ResNet50 Convolutional Neural Network (CNN) architecture. The banana data is divided into five ripeness categories: green, yellowish green, yellow, spotted yellow, and spotted brownish yellow. The model was trained with two approaches, with and without data augmentation, using two types of training algorithms (optimizers), namely Adam and SGD, as well as a k-fold cross-validation method to ensure accurate results. The results showed that the ResNet50 model produced the highest accuracy of 98% when trained using data augmentation and the Adam optimizer with a learning rate setting of 0.0001.

Classification Of Bougainvillea Flower Varieties Using Variant Of CNN: Resnet50 I Gede Agung Chandra Wijaya; I Gusti Agung Indrawan; I Nyoman Anom Fajaraditya; Ayu Gede Wildahlia; Ida Bagus Ary Indra Iswara
Indonesian Journal of Data and Science Vol. 6 No. 2 (2025): Indonesian Journal of Data and Science
Publisher : yocto brain

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56705/ijodas.v6i2.266

Bougainvillea is a tropical ornamental plant renowned for its vibrant colors and variety of cultivars, yet classifying its species remains challenging due to morphological similarities. This study aims to develop an automated classification system using the ResNet50 deep learning architecture to identify Bougainvillea flower varieties based on visual imagery. The dataset consists of 700 images from seven distinct classes, captured under natural lighting using a smartphone camera. The research process includes image preprocessing (resizing to 224x224 pixels), geometric data augmentation to increase dataset diversity, and evaluation using K-Fold Cross Validation to ensure robust model assessment. The model was trained using transfer learning, and its performance was compared between augmented and non-augmented datasets through evaluation metrics such as accuracy, precision, recall, and F1-score. The results show that augmentation significantly improved the model's performance, achieving an average accuracy of 99.67% on augmented data compared to 93.39% on non-augmented data. The augmented model also exhibited greater consistency across all folds, with several achieving perfect scores. These findings highlight that combining ResNet50 with transfer learning and image augmentation produces a highly accurate and reliable Bougainvillea classification system. This research contributes to the field of AI-based plant phenotyping and lays the groundwork for future applications in horticulture, biodiversity conservation, and education. Further development is recommended to explore larger and more diverse datasets, investigate advanced architectures such as EfficientNet or Vision Transformers, and build real-time mobile-based classification tools for practical field usage

Classification Of Organic And Inorganic Waste Using Resnet50 Qinantha, I Kadek Mahesa Chandra; Indrawan, I Gusti Agung; Putra, I Putu Satria Udyana; Aristamy, I Gusti Ayu Agung Mas; Willdahlia, Ayu Gede
Indonesian Journal of Data and Science Vol. 6 No. 2 (2025): Indonesian Journal of Data and Science
Publisher : yocto brain

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.56705/ijodas.v6i2.267

Waste generation, particularly from organic and inorganic sources, has become a growing environmental issue, especially in culturally unique regions like Bali where traditional offerings contribute to organic waste volumes. Despite regulations such as Gianyar Regency Regulation No. 76 of 2023 mandating source-level separation, on-ground implementation remains inconsistent due to low public awareness and operational limitations. This study addresses the challenge by developing an automated image-based classification system using the ResNet50 deep learning architecture to distinguish between organic and inorganic waste. A total of 200 images were collected 100 per class using smartphone cameras, and the dataset was expanded to 1,400 images through geometric data augmentation techniques such as rotation, flipping, and zooming. Images were resized to 224x224 pixels and evaluated using K-Fold Cross Validation to ensure model stability. The model was trained using transfer learning and tested under two conditions with and without augmentation while optimizing hyperparameters such as learning rates (0.0001 and 0.00001) and optimizers (Adam and SGD). The results demonstrate that augmentation significantly enhanced model performance, with the augmented model achieving an average accuracy of 99.25%, precision of 99.32%, recall of 99.25%, and F1-score of 99.25%, compared to 89.88% accuracy in the non-augmented model. These findings confirm that ResNet50, when combined with geometric augmentation and proper preprocessing, offers a robust, accurate, and scalable solution for waste classification tasks. This research contributes to the advancement of AI-driven environmental technologies and offers a potential framework for smart waste management systems, with future directions including real-time deployment, multi-class classification, and expansion to more diverse and real-world datasets.

Decision Support System for Sentiment Analysis of Youtube Comments on Government Policies Udayana, I Putu Agus Eka Darma; I Gusti Agung Indrawan; Putra, I Putu Dwi Guna Ambara
Journal of Computer Networks, Architecture and High Performance Computing Vol. 5 No. 1 (2023): Article Research Volume 5 Issue 1, January 2023
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v5i1.1999

Sentiment analysis is the process of classifying a text dataset as positive, negative or neutral. Youtube is one of the popular media used to provide responses to a problem. In the Jokowi era, infrastructure development was carried out massively and evenly, one of which was in Bali Province, namely the construction of the Mengwi-Gilimanuk Toll Road. The construction of the Mengwi-Gilimanuk Toll Road consumed a lot of people's agricultural land, which resulted in various pro and con responses from the community. From these problems, sentiment analysis is carried out to get community reviews related to the object being analyzed by utilizing algorithms to be able to classify opinions, in the construction of this system the naïve bayes algorithm is used with testing methods namely accuracy, precision, and recall. From the sentiment analysis conducted by utilizing 18 video links on YouTube with 701 comments, it produces positive sentiment as much as 50.64%, negative sentiment as much as 7.70% and neutral sentiment as much as 39.23%.

Comparison of Deep Learning Methods for Detecting Tuberculosis Through Chest X-Rays Udayana, I Putu Agus Eka Darma; Indrawan, I Gusti Agung; Prawira, I Made Karang Satria
Journal of Computer Networks, Architecture and High Performance Computing Vol. 6 No. 3 (2024): Articles Research Volume 6 Issue 3, July 2024
Publisher : Information Technology and Science (ITScience)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47709/cnahpc.v6i3.4345

Chronic diseases are the leading cause of death worldwide, accounting for 73% of deaths in 2020. Tuberculosis (TB), caused by the bacterium Mycobacterium tuberculosis, is one of these diseases and has a significant impact on countries with a high TB burden due to a lack of radiologists and medical equipment. Early diagnosis of TB is crucial but challenging because of its similarity to lung cancer and the shortage of radiologists. A semi-automatic TB detection system is needed to support medical diagnosis and improve public health services. Deep learning technology, such as Convolutional Neural Networks (CNN), offers an effective solution for disease diagnosis with high accuracy. This study compares deep learning methods using an 8-layer CNN and VGG-19, both enhanced with Histogram Equalization (HE) for improved image quality. The study utilizes chest X-ray images of normal lungs and TB-affected lungs from Kaggle. Model performance is evaluated using accuracy, precision, recall, and F1-score metrics. Results indicate that the VGG-19 model outperforms the 8-layer CNN across all evaluation metrics, achieving an accuracy of 72.00% compared to 65.00% for the 8-layer CNN. VGG-19 also demonstrates better precision, recall, and F1-score, making it a more suitable choice for TB detection with enhanced image quality.

Title

Found 23 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 23 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 23 Documents
Search