Claim Missing Document
Check
Articles

Found 27 Documents
Search

Community Assistance to Improve the Ability to Use Basic Unity and Virtual Reality Herawati Yusuf; Novie Theresia Br. Pasaribu; Erwani Merry Sartika; Evan Adrian Adhi; Reynaldy Felicius Gunawan; Richard Setiawan; Dion Melvern Siswanto; Vieri Candhya Wigayha; Nathaniel Pius Winata; Markus Tanubrata; Rainisa Maini Heryanto
Journal of Innovation and Community Engagement Vol. 4 No. 2 (2023)
Publisher : Faculty of Smart Technology and Engineering, Universitas Kristen Maranatha, Bandung, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/ice.v4i2.6502

Abstract

During the pandemic of Corona Virus Disease 2019 (COVID-19) that hit Indonesia, all activities were very limited. Electrical Engineering Study Program of Universitas Kristen Maranatha was motivated  to share knowledge especially in the field of science about "Basic Unity and Virtual Reality (VR)" to the public. This has been realized through community service activities pursued by students and lecturers. VR refers to computer simulations that create images of the world that appear in human senses, in the same way as humans perceive the real world in reality. This community service used the Participatory Action Research (PAR) approach. From the results of a survey of community service activities that have been carried out, it is concluded that the success in providing mentoring to improve the ability of the community to use Basic Unity and VR online, is shown by the results of an increase in the score from an average of 5.85 before the activity, to 7.58 (from a total score of 10) after the activity. The community service committee team's service during the activity was also considered very good and worthy of being maintained in subsequent activities. Meanwhile, the implementation of providing teaching materials, the readiness of the trainers, and the implementation of organizing activities online, still need to be re-evaluated and improved in the future for better results.
Online Workshop with Service Learning Method to Meet the Needs of Tinkercad Users from the Engineering Field Winda Halim; Erwani Merry Sartika; Novie Theresia Br. Pasaribu; Audyati Gany; Rainisa Maini Heryanto; Markus Tanubrata
Journal of Innovation and Community Engagement Vol. 4 No. 2 (2023)
Publisher : Faculty of Smart Technology and Engineering, Universitas Kristen Maranatha, Bandung, Indonesia

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.28932/ice.v4i2.6578

Abstract

The COVID-19 pandemic has greatly affected the learning process in practice, especially for people in the engineering field. Activities stopped and learning activities become very obstructed for a long time. Tinkercad is a simulator application that could be used as a learning tool or medium in engineering. The service learning method is used as a way of teaching and learning that links positive and meaningful action in society. Service learning uses demonstrations and experiments are used to transfer knowledge to society. A workshop elaborating the use of Tinkercad based on the service learning method was carried out online through community service activities to meet community needs in the engineering field. The workshop results showed that the service learning method was sufficient to help the community to meet the learning needs of engineering practices. A thorough demonstrations by instructors and experiments done by participants guided by assistants, produced a satisfactory result for more than 90% of the workshop participants. Periodic literacy provision was a follow-up to improve the abilities of all participants who have take part in workshop activities.
Prediksi Kuat Tekan Kolom Beton Persegi dengan Perkuatan FRP Menggunakan Machine Learning Anang Kristianto; Yosafat Aji Pranata; Novie Theresia Br. Pasaribu; Jasen Alfatama
RekaRacana: Jurnal Teknik Sipil Vol 11, No 2: Juli 2025
Publisher : Institut Teknologi Nasional, Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.26760/rekaracana.v11i2.158

Abstract

ABSTRAKAnalisis secara teoritis kuat tekan kolom beton persegi dengan perkuatan FRP masih mengandalkan pendekatan seperti yang digunakan untuk kolom lingkaran dengan modifikasi. Sesuai dengan kemajuan teknologi saat ini pendekatan menggunakan machine learning diharapakan dapat memberikan perspektif  berbeda sekaligus hasil yang lebih akurat. Tujuan dari penelitian ini adalah menggunakan beberapa teknik machine learning untuk memprediksi kuat tekan aksial kolom persegi yang diperkuat dengan FRP dengan berbagai radius kelengkungan pada sudutnya dan membandingkan metode machine learning dengan pendekatan analitik dan mekanik yang sudah ada. Tiga algoritma digunakan dalam pendekatan machine learning: Jaringan Syaraf Tiruan (JST), Categorical Boost (CatBoost), dan M5-Tree.  Ensemble Machine Learning membuat kinerja yang lebih baik dibandingkan dengan tiga model individual JST, CatBoost, dan M5-Tree dengan nilai R-squared sebesar 0,9557. Penggunaan machine learning untuk memprediksi dibandingkan dengan beberapa usulan teori tegangan terkekang, memberikan hasil yang paling mendekati dengan nilai R-squared sebesar 0,8276.Kata kunci: tegangan, kolom beton, FRP, prediksi, pembelajaran mesin ABSTRACTTheoretically analyzing the compressive strength of square concrete columns with FRP reinforcement still relies on the approach as used for circular columns with modifications. In accordance with current technological advances, machine learning approaches are expected to provide a different perspective as well as more accurate results. The objective of this research is to use several machine learning techniques to predict the axial compressive strength of square columns reinforced with FRP with various radii of curvature at the corners and compare the machine learning methods with existing analytical and mechanical approaches. Three algorithms were used in the machine learning approach: Artificial Neural Network (ANN), Categorical Boost (CatBoost), and M5-Tree.  The machine learning ensemble performed better than the three individual models JST, CatBoost, and M5-Tree with an R-squared value of 0.9557. The use of machine learning to predict, compared to some proposed confined stress theories, gave the closest results with an R-squared value of 0.8276.Keywords: stress, concrete column, FRP, prediction, machine learning
Petualangan Energi Angin Bersama Anak Rumah Singgah Erwani Merry Sartika; Meilan Jimmy Hasugian; Markus Tanubrata; Novie Theresia Br. Pasaribu; Herawati YS; Arvin Ezekiel Denri Utama; Dido Hardinanto Ginting; Jeffrey Christopher; Yehuda Njuah Sectio Cibro
Jurnal Atma Inovasia Vol. 6 No. 1 (2026)
Publisher : Lembaga Penelitian dan Pengabdian pada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24002/jai.v6i1.12212

Abstract

The wind energy adventure with children from the shelter home has become the focus of a community service initiative aimed at elementary school children in the Cibogo area. The need for alternative forms of learning to stimulate interest and engagement among children at the shelter inspired the application of science using a wind power car as the central theme of this community service program. The activity was conducted using an experimental method, training the children to observe, design tools, measure results, and draw conclusions from the objects they created—ranging from mechanical components to the drive system—through hands-on experimentation. The outcomes of this wind energy adventure had a positive impact on the children, the service team, and the surrounding community, fostering awareness of the importance of experiment-based education and providing valuable knowledge to all participants involved.
Kapsul Gravitasi sebagai Media Eksperimen Fisika Sederhana untuk Anak SD di Rumah Singgah Erwani Merry Sartika; Meilan Jimmy Hasugian; Markus Tanubrata; Novie Theresia Br. Pasaribu; Herawati YS; Arvin Ezekiel Denri Utama; Dido Hardinanto Ginting; Jeffrey Christopher; Yehuda Njuah Sectio Cibro
Jurnal Atma Inovasia Vol. 5 No. 6 (2025)
Publisher : Lembaga Penelitian dan Pengabdian pada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24002/jai.v5i6.12213

Abstract

The gravity capsule served as the main object of focus in a community service activity aimed at introducing science to children in the Cibogo Bandung shelter, specifically targeting first and second-grade elementary school students. The service learning method was employed to ensure the learning process remained simple and accessible, considering the young age of the participants. A basic experiment using the gravity capsule was conducted to introduce fundamental science concepts such as gravity, energy, and momentum. The outcome of this community service activity was considered successful, with a comprehension rate of 78.5%, based on responses to questions covering the materials delivered. However, improvements are still needed to enhance content retention among participants. Additionally, requests from the partner community for the program to be conducted regularly form the foundation for the sustainability of this initiative.
Pembelajaran STEM Berbasis Proyek Menggunakan Media Jembatan Berlampu Erwani Merry Sartika; Markus Tanubrata; Olga Catherina Pattipawaej; Rainisa Maini Heryanto; Meilan Jimmy Hasugian; Novie Theresia Br. Pasaribu; Dido Hardinanto Ginting; Gibran Qodri Nurfalah; Muhamad Dani Ardiansyah
Jurnal Atma Inovasia Vol. 6 No. 4 (2026)
Publisher : Lembaga Penelitian dan Pengabdian pada Masyarakat

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.24002/jai.v6i4.14215

Abstract

This community service activity aimed to enhance elementary school students’ understanding of bridge structural concepts and simple electrical circuits through individual project-based STEM learning. The activity was conducted with Grade 3–4 students in the Cibogo area, utilizing simple materials such as popsicle sticks, LEDs, wires, batteries, and switches as learning media. The methodology integrated community-based research, participatory learning, and action research approaches, in which the activities were designed based on the needs of the target community, actively engaged students in the learning process, and incorporated reflective practices for continuous improvement. Evaluation was carried out using a questionnaire to assess students’ understanding of bridge structures, simple electrical circuits, and their responses to the activity. The evaluation results indicated that most students were able to comprehend the fundamental concepts introduced and showed positive responses to the project-based learning activities (85% for bridge construction and 78% for electrical circuits). In addition to improved conceptual understanding, students demonstrated high levels of interest and enthusiasm throughout the activity. It can be concluded that this activity was effective in providing meaningful and contextual learning experiences. Future implementations may be further developed by strengthening introductory materials and diversifying project models to enhance the quality and sustainability of STEM learning in elementary school environments.
Comparison of Filter and SHAP Feature Selection for ECG-based Atrial Fibrillation Classification Novie Theresia Pasaribu; Elizabeth Fabiola Wijaya; Che Wei Lin
Jambura Journal of Electrical and Electronics Engineering Vol 8, No 2 (2026): Juli - Desember 2026
Publisher : Electrical Engineering Department Faculty of Engineering State University of Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjeee.v8i2.39633

Abstract

Atrial Fibrillation (AF) is a type of arrhythmia whose prevalence continues to rise globally and can lead to serious complications such as stroke and heart attack. Early detection based on ECG signals is therefore crucial. This study compares three feature selection methods, namely Pearson Correlation (PC), Mutual Information (MI), and Shapley Additive Explanations (SHAP), for classifying AF from 2-lead ECG signals using XGBoost. The dataset used was the MIT-BIH Atrial Fibrillation Database, comprising 23 ECG recordings, yielding 34,312 10-second data segments. Preprocessing used Stationary Wavelet Transform (SWT) and Min-Max normalisation. A total of 50 features were extracted. Each method was tested on 4 feature subset sizes (5, 10, 15, and 20 features). The model was optimised using GridSearchCV with 5-fold Stratified Cross-Validation. Results showed that SHAP outperformed PC and MI in subsets of 10, 15, and 20 features. SHAP with 20 features achieved the highest performance (97.87% accuracy; F1 score 96.78%; ROC-AUC 0.9971), while the 15-feature SHAP offered the best performance–dimensionality trade-offs: 97.84% accuracy, 97.85% precision, 95.62% recall, 96.72% F1-score, and 0.9965 ROC-AUC, with a 70% dimensional reduction (0.03% below SHAP-20 on accuracy, with higher precision). SHAP's superiority over PC and MI was statistically significant (McNemar and DeLong tests, p 0.05) in subsets of 15 and 20 features; SHAP with 20 features achieved a ROC-AUC that did not differ significantly from the baseline of 50 features (DeLong test, p = 0.4997).