Garuda - Garba Rujukan Digital

Sentiment Analysis Of Bali's Rapid Tourism Growth Using IndoBERT With InSet Lexicon Automatic Labelling Ni Ketut Artini Artalia; I Made Dendi Maysanjaya; Putu Hendra Suputra
KARMAPATI (Kumpulan Artikel Mahasiswa Pendidikan Teknik Informatika) Vol. 15 No. 1 (2026): Karmapati Vol 15 No 1 Tahun 2026
Publisher : Universitas Pendidikan Ganesha

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.23887/karmapati.v15i1.111839

Bali’s rapid tourism growth has generated diverse public opinions, particularly on TikTok, where comment sections reflect debates and societal concerns. The absence of an automated sentiment monitoring system may delay policy responses and potentially intensify social conflict. Therefore, sentiment analysis is required to understand public perception comprehensively. This study conducts sentiment analysis on TikTok comments related to Bali’s tourism growth using a transformer-based approach. IndoBERT, a pre-trained model for Indonesian-language processing, is employed due to its superior performance compared to traditional machine learning methods. Sentiment labelling is performed using the Indonesian Sentiment Lexicon (InSet Lexicon), which provides an efficient and consistent automatic labelling alternative to manual annotation. Model performance is evaluated through two experimental scenarios: IndoBERT with InSet Lexicon automatic labelling, and IndoBERT combined with InSet Lexicon and Random Over Sampling (ROS). The results show that applying ROS significantly improves classification performance by addressing class imbalance, resulting in higher accuracy, precision, recall, and F1-score. The analysis indicates that public sentiment toward Bali’s tourism growth is predominantly neutral, suggesting that most user express opinions in an informative and descriptive manner rather than strong emotional responses. Overall, the integration of IndoBERT, InSet Lexicon automatic labelling, and Random Over Sampling is proven to be an effective approach for Indonesian sentiment analysis, particularly in examining socially informative issues such as tourism development in Bali.

Automatic 3D Cranial Landmark Positioning based onSurface Curvature Feature using Machine Learning Suputra, Putu Hendra; Sensusiati, Anggraini Dwi; Artaria, Myrtati Dyah; Verkerke, Gijsbertus Jacob; Yuniarno, Eko Mulyanto; Purnama, I Ketut Eddy
Knowledge Engineering and Data Science
Publisher : citeus

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

Cranial anthropometric reference points (landmarks) play an important role in craniofacial reconstruction and identification. Knowledge to detect the position of landmarks is critical. This work aims to locate landmarks automatically. Landmarks positioning using Surface Curvature Feature (SCF) is inspired by conventional methods of finding landmarks based on morphometrical features. Each cranial landmark has a unique shape. With the appropriate 3D descriptors, the computer can draw associations between shapes and landmarks using machine learning. The challenge in classification and detection in three-dimensional space is to determine the model and data representation. Using three-dimensional raw data in machine learning is a serious volumetric issue. This work uses the Surface Curvature Feature as a three-dimensional descriptor. It extracts the local surface curvature shape into a projection sequential value (depth). A machine learning method is developed to determine the position of landmarks based on local surface shape characteristics. Classification is carried out from the top-n prediction probabilities for each landmark class, from a set of predictions, then filtered to get pinpoint accuracy. The landmark prediction points are hypothetically clustered in a particular area, so a cluster-based filter is appropriate to isolate them. The learning model successfully detected the landmarks, with the average distance between the prediction points and the ground truth being 0.0326 normalized units. The cluster-based filter is implemented to increase accuracy compared to the ground truth. Thus, SCF is suitable as a 3D descriptor of cranial landmarks.

Pendidikan Matematika Inklusif untuk Siswa Tunarungu: Pengembangan Model YOLO Ringan untuk Deteksi Operator Aritmetika Dua Tangan Putu Nanditha Nayottama; Made Gede Sunarya; Putu Hendra Suputra
JURNAL ILMIAH SAINS TEKNOLOGI DAN INFORMASI Vol. 4 No. 3 (2026): Juli : Jurnal Ilmiah Sains Teknologi dan Informasi
Publisher : CV. ALIM'SPUBLISHING

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.59024/jiti.v4i3.2165

This study presents a real-time recognition system for two-handed arithmetic operators (Plus, Minus, Multiply, Divide) to support inclusive mathematics education for deaf students. Utilizing the YOLO26 architecture, the research evaluates four configurations—Nano and Small variants at 640px and 720px resolutions—trained on a newly developed private dataset of 2,828 images. Experimental results demonstrate exceptional accuracy, with all configurations achieving mAP50 scores above 0.99. The YOLO26-s variant at 720px provided the highest precision (0.9842), while benchmarking on a local GTX 1650 GPU confirmed real-time viability with frame rates reaching up to 88 FPS. Even when deployed on a mobile CPU (i5-12500H), the lightweight Nano models maintained a functional execution speed of 23 FPS. These findings confirm that optimized deep learning models can accurately interpret complex mathematical gestures on consumer-grade hardware. Ultimately, this lightweight framework successfully bypasses heavy skeletal tracking overhead, providing a highly robust and accessible foundation for interactive digital educational media.

Klasifikasi Jenis-Jenis Ikan Konsumsi di Pasar Banyuasri Menggunakan Convolutional Neural Network dengan Arsitektur VGG-16 dan MobileNetV2 Gusti Nyoman Satya Sudharsan; Putu Hendra Suputra; Ni Wayan Kertiasih
JURNAL PENELITIAN SISTEM INFORMASI (JPSI) Vol. 4 No. 3 (2026): Agustus : JURNAL PENELITIAN SISTEM INFORMASI
Publisher : Institut Teknologi dan Bisnis (ITB) Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54066/jpsi.v4i3.4477

Abstract. The diversity of consumption fish species in traditional markets often creates difficulties in the identification process, particularly for species with similar morphological characteristics. This study aims to compare the performance of two Convolutional Neural Network (CNN) architectures, namely VGG-16 and MobileNetV2, in classifying 15 species of consumption fish sold at Banyuasri Market, Singaraja. The dataset consisted of 750 images, which were divided into 600 training images and 150 testing images. Furthermore, the training dataset was split into 480 training images and 120 validation images. During the training process, on-the-fly image augmentation was applied using rotation, width and height shifting, zoom, shear, horizontal flipping, and brightness adjustment to improve the model's generalization capability. The training process was conducted in two stages: an initial training phase of 20 epochs followed by a fine-tuning phase of 25 epochs. The experimental results showed that MobileNetV2 outperformed VGG-16, achieving an accuracy of 90.67%, precision of 91.28%, recall of 90.67%, and an F1-score of 90.66%, while VGG-16 achieved an accuracy of 88.00%, precision of 89.03%, recall of 88.00%, and an F1-score of 87.33%. Based on these findings, MobileNetV2 is considered more effective and computationally efficient for consumption fish classification and has strong potential to serve as the foundation for developing web-based and mobile-based fish identification systems.

Analisis Sentimen Opini Publik terhadap Inovasi Teknologi pada Administrasi Publik Indonesia Menggunakan Metode IndoBERT Komang Tya Vinandita Kusuma Dewi; Putu Hendra Suputra; I Nyoman Tri Anindia Putra
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 5 No. 2 (2026): Mei-Juli
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v5i2.10149

Transformasi digital dalam administrasi publik di Indonesia mendorong pemerintah untuk meningkatkan kualitas pelayanan melalui penerapan berbagai inovasi teknologi. Namun, implementasi layanan digital pemerintah masih menghasilkan persepsi yang beragam di masyarakat yang dapat dilihat melalui opini publik pada media sosial. Penelitian ini bertujuan untuk menganalisis sentimen opini publik terhadap inovasi teknologi pada administrasi publik Indonesia menggunakan metode IndoBERT, mengidentifikasi topik dominan yang dibahas masyarakat menggunakan Latent Dirichlet Allocation (LDA), serta menyajikan hasil analisis dalam bentuk dashboard interaktif. Penelitian ini menggunakan pendekatan kuantitatif dengan kerangka kerja Knowledge Discovery in Database (KDD) yang terdiri dari tahapan data selection, preprocessing, transformation, data mining, dan evaluation. Data diperoleh dari media sosial X dan TikTok sebanyak 11.787 data opini yang kemudian melalui proses seleksi dan preprocessing sehingga menghasilkan 5.541 data yang digunakan untuk analisis. Klasifikasi sentimen dilakukan menggunakan model IndoBERT melalui proses fine-tuning dengan beberapa kombinasi hyperparameter. Evaluasi model dilakukan menggunakan confusion matrix, akurasi, presisi, recall, F1-Score, dan 5-Fold Cross Validation. Hasil penelitian menunjukkan bahwa konfigurasi terbaik diperoleh pada epoch 4 dengan learning rate 2e-5 yang menghasilkan akurasi sebesar 75% dan F1-Score sebesar 74%. Selain itu, hasil topic modeling menghasilkan lima topik utama pada masing-masing kelas sentimen yang menggambarkan isu dominan terkait penerapan inovasi teknologi administrasi publik. Seluruh hasil analisis divisualisasikan dalam dashboard berbasis website untuk membantu pengguna memahami pola persepsi masyarakat dan mendukung evaluasi layanan digital pemerintah.

Analisis Sentimen terhadap Perencanaan Pemilu Elektronik di Indonesia Menggunakan Pendekatan Soft Voting Ensemble I Putu Dennis Prana Arta; Putu Hendra Suputra; I Gusti Ayu Agung Diatri Indradewi
RIGGS: Journal of Artificial Intelligence and Digital Business Vol. 5 No. 2 (2026): Mei-Juli
Publisher : Prodi Bisnis Digital Universitas Pahlawan Tuanku Tambusai

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31004/riggs.v5i2.10298

Pemilu elektronik merupakan salah satu inovasi yang berpotensi meningkatkan efisiensi, transparansi, dan akurasi dalam penyelenggaraan pemilihan umum. Namun, implementasinya masih menimbulkan beragam tanggapan dari masyarakat yang banyak disampaikan melalui media sosial. Penelitian ini bertujuan untuk menganalisis sentimen masyarakat terhadap implementasi pemilu elektronik di Indonesia menggunakan metode Soft Voting Ensemble. Penelitian menerapkan kerangka kerja CRISP-DM dengan memanfaatkan data yang diperoleh dari platform X dan TikTok. Dari 9.715 data yang berhasil dikumpulkan, diperoleh 3.979 data relevan setelah proses seleksi, yang kemudian dilabeli secara manual oleh tiga ahli bahasa Indonesia menjadi 1.446 sentimen positif dan 2.533 sentimen negatif. Tahapan penelitian meliputi preprocessing, pembobotan TF-IDF, penanganan ketidakseimbangan data menggunakan Borderline-SMOTE, klasifikasi menggunakan Logistic Regression, Random Forest, dan Soft Voting Ensemble, serta evaluasi menggunakan confusion matrix dan 5-Fold Cross Validation. Hasil penelitian menunjukkan bahwa model Soft Voting Ensemble dengan optimasi awal dan Borderline-SMOTE menghasilkan performa terbaik dengan akurasi sebesar 80,78%. Selain itu, analisis topik menggunakan Latent Dirichlet Allocation (LDA) berhasil mengidentifikasi topik-topik dominan yang memengaruhi sentimen positif dan negatif masyarakat terhadap pemilu elektronik. Hasil penelitian juga diimplementasikan dalam bentuk dashboard interaktif untuk mendukung visualisasi dan pemanfaatan informasi oleh pemangku kepentingan. Temuan penelitian menunjukkan bahwa pendekatan Soft Voting Ensemble efektif dalam menganalisis sentimen publik terhadap implementasi pemilu elektronik di Indonesia.

Klasifikasi Jenis-Jenis Ikan Konsumsi di Pasar Banyuasri Menggunakan Convolutional Neural Network dengan Arsitektur VGG-16 dan MobileNetV2 Gusti Nyoman Satya Sudharsan; Putu Hendra Suputra; Ni Wayan Kertiasih
JURNAL PENELITIAN SISTEM INFORMASI (JPSI) Vol. 4 No. 3 (2026): Agustus : JURNAL PENELITIAN SISTEM INFORMASI
Publisher : Institut Teknologi dan Bisnis (ITB) Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.54066/jpsi.v4i3.4477

Abstract. The diversity of consumption fish species in traditional markets often creates difficulties in the identification process, particularly for species with similar morphological characteristics. This study aims to compare the performance of two Convolutional Neural Network (CNN) architectures, namely VGG-16 and MobileNetV2, in classifying 15 species of consumption fish sold at Banyuasri Market, Singaraja. The dataset consisted of 750 images, which were divided into 600 training images and 150 testing images. Furthermore, the training dataset was split into 480 training images and 120 validation images. During the training process, on-the-fly image augmentation was applied using rotation, width and height shifting, zoom, shear, horizontal flipping, and brightness adjustment to improve the model's generalization capability. The training process was conducted in two stages: an initial training phase of 20 epochs followed by a fine-tuning phase of 25 epochs. The experimental results showed that MobileNetV2 outperformed VGG-16, achieving an accuracy of 90.67%, precision of 91.28%, recall of 90.67%, and an F1-score of 90.66%, while VGG-16 achieved an accuracy of 88.00%, precision of 89.03%, recall of 88.00%, and an F1-score of 87.33%. Based on these findings, MobileNetV2 is considered more effective and computationally efficient for consumption fish classification and has strong potential to serve as the foundation for developing web-based and mobile-based fish identification systems.

Rekognisi Bahasa Isyarat Indonesia (BISINDO) Berbasis Long-Short Term Memory Dengan Optimasi Sampling Frame Gunaya, I Wayan; Kesiman, Made Windu Antara; Suputra, Putu Hendra
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 13 No 4: Agustus 2026
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.134

Bahasa Isyarat Indonesia atau BISINDO merupakan bahasa isyarat yang lahir dari komunitas Tuli yang ada di Indonesia. Gerakan tangan, tubuh, hingga ekspresi wajah merupakan bagian terpenting dalam penggunaan bahasa isyarat. Pada penelitian ini, dilakukan proses rekognisi gerakan bahasa isyarat berdasarkan gerakan tangan, tubuh, dan wajah dengan menggunakan metode LSTM. Dataset yang digunakan merupakan dataset primer yang terdiri dari 100 kosa isyarat yang dibuat berdasarkan dialek Denpasar, Provinsi Bali. Penelitian ini tidak hanya berfokus pada proses rekognisi, tetapi juga mengevaluasi pengaruh variasi jumlah frame terhadap performa model melalui penerapan metode sampling frame. Pemilihan jumlah frame dilakukan dengan memilih representasi frame sebanyak 15, 20, 25, dan 30 frame. Keempat jumlah sampling frame tersebut kemudian dilatih dan diuji pada model LSTM. Hasil penelitian ini menunjukkan hasil sampling frame optimal adalah pada 30 frames yang memperoleh hasil pelatihan dengan nilai loss validasi terkecil pada 7,8% dan hasil pengujian WER terbaik pada 13% jika dibandingkan dengan hasil sampling frame lainnya. Berdasarkan hasil penelitian ini, diharapkan model LSTM yang telah diperoleh dan diimplementasikan ke dalam prototipe sistem dapat digunakan oleh masyarakat umum dalam menjembatani komunikasi masyarakat umum dengan komunitas Tuli serta oleh peneliti selanjutnya dalam pengembangan lebih lanjut proses rekognisi gerakan Bahasa Isyarat Indonesia. Abstract Indonesian Sign Language or BISINDO is a natural sign language that emerged from the Deaf community in Indonesia. Hand movements, body posture, and facial expressions play essential roles in conveying meaning within sign language communication. In this study, sign language gesture recognition is performed based on hand, body, and facial movements using the Long Short-Term Memory (LSTM) method. The dataset used is a primary dataset consisting of 100 sign vocabulary items constructed based on the Denpasar dialect in Bali Province. This study does not solely focus on the recognition process, but also evaluates the effect of varying the number of frames on model performance through the application of a frame sampling method. The selection of frame quantities was conducted by choosing representative samples of 15, 20, 25, and 30 frames. These four sampling configurations were subsequently trained and tested using an LSTM model. The results of this study indicate that the optimal frame sampling configuration is 30 frames, which achieved the lowest validation loss of 7.8% during training and the best Word Error Rate (WER) of 13% during testing, compared to the other sampling configurations. Based on these findings, it is expected that the LSTM model developed and implemented within a prototype system can be utilized by the general public to help bridge communication between the general public and the Deaf community and by future researchers for the further development of Indonesian Sign Language gesture recognition processes.

Title

Found 28 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 28 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 28 Documents
Search