Article Per Year (5 Year)
p-Index From 2021 - 2026
3.673
P-Index
This Author published in this journals
All Journal Jurnal Informatika Jurnal Teknologi Informasi dan Ilmu Komputer Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) Jurnal Informatika JURNAL MEDIA INFORMATIKA BUDIDARMA CogITo Smart Journal SELAPARANG: Jurnal Pengabdian Masyarakat Berkemajuan JURNAL TEKNIK INFORMATIKA DAN SISTEM INFORMASI Infotekmesin JISA (Jurnal Informatika dan Sains) IJIIS: International Journal of Informatics and Information Systems Abdimasku : Jurnal Pengabdian Masyarakat SEMAR Teknika Journal of Applied Data Sciences Jurnal Pendidikan dan Teknologi Indonesia Nusantara Hasana Journal Society : Jurnal Pengabdian dan Pemberdayaan Masyarakat International Journal for Applied Information Management
Chyntia Raras Ajeng Widiawati
Universitas Amikom Purwokerto
Agriculture, Biological Sciences & Forestry Humanities Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance Education Electrical & Electronics Engineering Engineering Environmental Science Health Professions Industrial & Manufacturing Engineering Languange, Linguistic, Communication & Media Law, Crime, Criminology & Criminal Justice Mechanical Engineering Medicine & Pharmacology Nursing Public Health Social Sciences Other

Published : 25 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 25 Documents
Search

 1   2   3 
Enhancing Minority Class Prediction in Wearable Sensor-Based Activity Recognition Using SMOTE Oversampling Sarmini; Widiawati, Chyntia Raras Ajeng; Yunita, Ika Romadoni
International Journal for Applied Information Management Vol. 5 No. 1 (2025): Regular Issue: April 2025
Publisher : Bright Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/ijaim.v5i1.95

Abstract

Wearable sensor-based activity recognition has become increasingly important in various domains, particularly healthcare and sports. However, a significant challenge in this field is the issue of class imbalance, where minority activity classes are underrepresented compared to majority classes in datasets. This imbalance leads to biased classifiers that struggle to accurately identify rare but critical activities, which is especially problematic in health monitoring scenarios. This study evaluates the effectiveness of the Synthetic Minority Oversampling Technique (SMOTE) to address class imbalance in the mHealth dataset, which contains multi-sensor data from wearable devices placed on the chest, left ankle, and right lower arm. We employ the XGBoost classifier combined with SMOTE oversampling to improve recognition performance for minority classes. Model evaluation is conducted using precision, recall, F1-score, Area Under the Precision-Recall Curve (AUC-PR), ROC curve, and calibration analysis. The results demonstrate that applying SMOTE improves minority class recall from 0.75 to 0.85 and F1-score from 0.796 to 0.865, despite a slight decrease in overall accuracy from 97% to 96.5%. The AUC-PR also increases from 0.81 to 0.88, indicating a better balance in detecting minority and majority classes. Calibration curves reveal that probability estimates still require refinement to be more reliable for decision-making. This study confirms the efficacy of SMOTE in mitigating class imbalance in wearable sensor-based activity recognition and provides valuable insights for developing more accurate and fair health monitoring systems.
Determinants of Consumption Behavior Among the Millennial Generation Saputra, Aina Aldi; Sarmini, Sarmini; Widiawati, Chyntia Raras Ajeng; Yunita, Ika Romadoni
International Journal of Informatics and Information Systems Vol 7, No 3: September 2024
Publisher : International Journal of Informatics and Information Systems

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/ijiis.v7i3.216

Abstract

This study examines the factors influencing consumption behavior among the millennial generation, emphasizing the effects of family size, education level, and income on food, non-food, and total household expenditure. As digitalization and demographic shifts continue to redefine modern lifestyles, understanding millennial consumption patterns offers valuable insights into changing welfare dynamics and economic structures. Employing a quantitative associative approach, data were collected from 120 millennial households through structured questionnaires and interviews, complemented by secondary data from the Central Statistics Agency (BPS). Multiple linear regression analysis was used to evaluate both simultaneous and partial relationships among variables, while descriptive statistics were applied to illustrate the respondents’ socioeconomic characteristics. The findings show that family size, education, and income collectively have a significant influence on consumption across all categories. Partially, family size and income significantly affect food-related spending, whereas education does not exhibit a notable impact in this segment. In contrast, for non-food and total consumption, all three variables display a positive and significant relationship, suggesting that higher income and education levels encourage more diversified expenditures. Moreover, non-food consumption (57.19%) surpasses food consumption (42.81%), supporting Engel’s Law and indicating improved living standards alongside a shift toward lifestyle diversification. Nonetheless, the proportion of non-food expenditure remains moderate, reflecting cautious financial behavior amid lingering post-pandemic income constraints. These findings align with Keynesian and Life-Cycle consumption theories, illustrating how income stability, education, and life-stage factors shape millennial consumption decisions. Overall, this study underscores the evolving nature of millennial households toward technology-driven, experience-based, yet financially mindful consumption patterns, providing implications for policymakers and businesses to enhance income resilience, digital literacy, and sustainable consumption growth in the digital economy.
Pengaruh Dataset terhadap Performa Convolutional Neural Network pada Klasifikasi X-Ray Pasien Covid-19 Widiawati, Chyntia Raras Ajeng
Jurnal Teknologi Informasi dan Ilmu Komputer Vol 9 No 6: Desember 2022
Publisher : Fakultas Ilmu Komputer, Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25126/jtiik.2022965645

Abstract

Identifikasi pasien COVID-19 di Indonesia merupakan sebuah permasalahan yang harus diatasi. Identifikasi tersebut sebaiknya bisa lebih mudah dan cepat, sehingga deteksi dini pasien COVID-19 merupakan hal yang harus diperhatikan. Saat ini deteksi pasien COVID-19 bergantung pada Swab Test (RT-PCR) dan Rapid Test (Tes Antibodi), padahal Rapid Test tidak bisa memberikan tingkat akurasi yang tinggi sedangkan Swab Test memiliki biaya yang mahal. Salah satu solusi untuk membantu deteksi dini pasien COVID-19 adalah dengan memanfaatkan citra X-Ray paru dari pasien. Algoritma Convolutional Neural Network (CNN) adalah salah satu algoritma popular dengan performa yang sangat baik pada klasifikasi citra X-Ray pasien COVID-19. Walaupun CNN memiliki performa yang baik, keberhasilan suatu algoritma sangat bergantung dengan kualitas dataset yang digunakan. Selain itu citra X-Ray sangat bergantung terhadap pencahayaan di proses pengambilan gambar. Untuk itu perlu analisa pengaruh dataset terhadap performa model CNN yang digunakan. Penelitian ini bertujuan untuk melihat pengaruh kualitas dataset dan jumlah dataset terhadap performa CNN pada klasifikasi X-Ray pasien COVID-19. Dari eksperimen yang dilakukan terhadap dataset yang ada, CNN dapat mencapai akurasi sebesar 89,83%, sensitivitas sebesar 84,14%, spesifisitas sebesar 92,14%, PPV sebesar 71,35%, NPV sebesar 95,09 dan F1-score sebesar 76,10%. Sehingga dapat disimpulkan bahwa CNN memiliki performa yang baik dalam melakukan klasifikasi citra X-Ray pasien COVID-19, meskipun hasil tersebut lebih rendah dibandingkan dengan performa CNN terhadap dataset dengan jumlah dan kualitas citra yang lebih baik. Hal tersebut menunjukkan bahwa kualitas dan jumlah dataset sangat berpengaruh pada performa CNN dalam melakukan proses klasifikasi X-Ray.
GAN-Enhanced Radial Basis Function Networks for Improved Landslide Susceptibility Mapping Widiawati, Chyntia Raras Ajeng; Maulita, Ika; Purwati, Yuli; Wahid, Arif Mu'amar
Journal of Applied Data Sciences Vol 7, No 1: January 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i1.1035

Abstract

Landslide susceptibility modeling is a critical task for disaster mitigation, yet it is frequently undermined by a severe class imbalance inherent in landslide datasets, where non-landslide instances vastly outnumber actual landslide events. This imbalance leads to biased machine learning models with poor predictive power for the minority (landslide) class, resulting in unreliable hazard maps. This study, focusing on the high-risk area of Malang Regency, Indonesia, addresses this challenge by proposing an innovative framework that integrates a Generative Adversarial Network (GAN) for synthetic data augmentation with a Radial Basis Function Network (RBFN) for classification. A highly imbalanced dataset with a 1:10 ratio of landslide to non-landslide points was constructed to establish a realistic baseline. On this data, the RBFN model, while theoretically powerful for capturing non-linear relationships, failed completely, achieving a Recall of 0.00 for the landslide class. The novelty of this research lies in the specific application of a GAN, trained for 15,000 epochs, to generate high-fidelity synthetic landslide data, thereby creating a perfectly balanced training set. After retraining on this augmented data and undergoing a systematic hyperparameter tuning process, the RBFN’s performance was dramatically transformed. The optimized model achieved an F1-Score of 0.9333 and a Recall of 0.8750, elevating its performance from total failure to a level competitive with the robust Random Forest benchmark. This work validates that the integrated GAN-RBFN approach is a highly effective methodology for overcoming the data imbalance problem in geospatial hazard modeling. By turning a previously unreliable classifier into a powerful predictive tool, this method has significant practical implications for developing more accurate landslide susceptibility maps, which are crucial for informed spatial planning and enhancing early warning systems.
Automatic RoI dan Active Contour untuk Deteksi Penggunaan Helm pada Pengendara Sepeda Motor widiawati, chyntia raras ajeng
JISA(Jurnal Informatika dan Sains) Vol 2, No 2 (2019): JISA(Jurnal Informatika dan Sains)
Publisher : Universitas Trilogi

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31326/jisa.v2i2.490

Abstract

Based on data from the Central Statistics Agency, motorcycle accidents are the most common accidents and many contribute to the death rate in traffic accidents. Some cases of deaths in motorcycle accidents are caused by riders not wearing helmets. Monitoring via CCTV video has been done but it takes a long time so we need another solution to be more effective. Some techniques have been carried out including detection of the use of helmets on motorcyclists by using digital image processing. Some previous studies on these cases experienced obstacles such as overlapping images in the identification process. This study aims to develop a detection method that focuses on the segmentation stage to produce a better segmentation image. The method used in this study is Automatic RoI and Active Contour at the segmentation stage which is then classified using the Multilayer Perceptron classifier. The results obtained give an accuracy value of 72.97%, a sensitivity of 76.19% and a specificity of 68.75%.
Co-Authors Achmad Masruri Adiya, Az Zahra Dwi Nur Alif Yahya Syafa'at Anjar Wati Anugerah Bagus Wijaya Atmaja Jalu Narendra Kisma Aulia Shafira Tri Damayanti Bambang Pilu Hartato Bela Tyara Nur Fajrina Danty Kusumaningtyas Delia Oktaviana Azzahra Dila Putri Sabil Fandy Setyo Utomo Febrianti, Diah Ratna Ganang Eko Saputro Gina Cahya Utami Ika Maulita Ika Romadoni Yunita Ilham, Rifqi Arifin Indra Alan Nugroho Irfan Rifai Aziz Lisa Nurazizah Luthfi Khaerunnisa Pungkas Subarkah Rahayu, Mia Setya Retno Waluyo Rizky Arsyansyah Rujianto Eko Saputro Saputra, Aina Aldi Sarmini Sarmini Sarmini Sarmini Sarmini Sri Hartini Suliswaningsih Tegar Insani Wahid, Arif Mu'amar Wiga Maulana Baihaqi Wiga Maulana Baihaqi Yuli Purwati Yuliana, Dwi Yunita, Ika Romadoni