Universitas Ahmad Dahlan Journal
Not a member yet
5744 research outputs found
Sort by
Diabetes Mellitus Classification Using CNN-Based Plantar Thermogram Analysis
Full text linkDiabetes Mellitus (DM) is a chronic metabolic disorder that often causes serious complications, including neuropathy and lower extremity disorders, which impact the quality of life of patients. Early detection of DM is a major challenge due to limited data and the complexity of image analysis. This study proposes a plantar thermogram image-based approach to support non-invasive diagnosis of DM through the development of a Convolutional Neural Network (CNN)-based model and machine learning techniques. This model integrates data augmentation techniques, such as rotation, flip, and zoom, to improve image variation and model robustness. Two CNN architectures, InceptionV3 and ResNet-50, are used in the training process, followed by feature selection using the Chi-Square method and classification using the Random Forest algorithm. The results showed that the proposed model achieved the best performance with accuracy, F1-score, precision, recall, and AUC (Area Under Curve) of 99.6% each. This approach makes a significant contribution by showing improvement compared to previous methods, while opening up opportunities for the development of more efficient clinical applications in early detection and monitoring of DM
Unveiling the Growth and Development of Electrical, Computer, and Informatics Engineering Education: A Bibliometric Perspective
Full text linkThis study presents a bibliometric analysis of research trends in Electrical, Computer, and Informatics Engineering Education from 2015 to 2023, focusing on the integration of emerging technologies such as AI, IoT, and e-learning platforms. Data was extracted from the Scopus database, and analysis was conducted using co-occurrence analysis and citation network mapping. The study identifies key research themes, such as the shift towards active learning methodologies (e.g., problem-based learning and gamification) and the growing emphasis on technology-driven curricula. Findings show a significant rise in research output, particularly during the COVID-19 pandemic, with IEEE journals dominating publications in the field. The results highlight the transformative role of digital tools in engineering education and the challenges of balancing technological integration with traditional teaching methods. This research offers insights into the evolving landscape of engineering education and provides recommendations for future research directions
The influence of the brain-based learning (BBL) model assisted by multimedia phet simulation on critical thinking ability
No full textMathematics learning in elementary schools is still dominated by conventional methods that do not foster the development of critical thinking skills. This challenge is relevant to 21st-century learning and the Independent Curriculum, which demands the integration of technology and high-level thinking skills. This research aims to analyze the effect of Brain-Based Learning (BBL) supported by PhET Simulations on the critical thinking abilities of fifth-grade students in Magelang City. The research approach uses a quasi-experimental design with a nonequivalent control group. The research instrument is an essay test to measure critical thinking abilities. The results of the analysis show that the BBL model assisted by PhET Simulation provides a significant improvement in critical thinking abilities. Data testing uses the t-test and N-Gain to prove the hypothesis. The t-test produces a significance value of 0.00 < 0.05, indicating a significant effect of using BBL assisted by the PhET Simulation. This finding is strengthened by the average N-Gain value of 0.5002, which falls within the medium category, indicating that brain-based work learning is quite effective in improving students' critical thinking abilities. Integration of BBL and PhET Simulations proved effective in supporting 21st-century skills. Implementing this model can provide learning experiences that support optimal brain function through technology, making it relevant for developing 21st-century skills
Navigating the regulatory landscape: Combating corruption, cryptocurrency crime, and illicit finance through global coordination
Full text linkIntroduction to the Problem: This article examines the U.S. strategy for countering corruption and the increasing challenges of money laundering involving cryptocurrencies in a globalized financial ecosystem. As digital assets gain legitimacy, they have simultaneously become tools for illicit finance, prompting the need for coordinated global regulatory efforts. The United States, home to the world’s largest crypto exchanges and a leading jurisdiction for asset seizures, has developed a comprehensive Five-Pillar Strategy emphasizing global coordination and institutional strengthening.
Purpose/Objective Study: This study analyzes how U.S. policy frameworks, including those under the Commodity Futures Trading Commission (CFTC), Financial Crimes Enforcement Network (FinCEN), and Dodd-Frank Act, respond to transnational threats of corruption, crypto-related crime, and illicit finance. It assesses how these measures promote transparency and shape international cooperation mechanisms.
Design/Methodology/Approach: Using a mixed-method legal approach grounded in methodological pluralism, this research integrates normative legal analysis, legal sociology, and neoliberal institutionalism to evaluate the adaptive capacity of global coordination in addressing crypto-related financial crimes.
Findings: The study finds that effective responses to crypto-based corruption require not only domestic policy coherence but also institutionalized multilateral coordination anchored in international regimes such as the Financial Action Task Force (FATF), the UN Convention against Corruption (UNCAC), and the OECD’s Crypto-Asset Reporting Framework. The U.S. Five-Pillar Strategy strengthens transparency through beneficial ownership reporting, enhances the detection of illicit transactions via FinCEN and CFTC oversight, and reinforces cross-border collaboration through FATF and UNCAC partnerships. These frameworks collectively represent a pragmatic application of neoliberal institutionalism (where institutions mitigate the risks of an anarchic financial order) and sociological jurisprudence, which treats law as a dynamic tool of social engineering. However, gaps persist in enforcement harmonization and data-sharing, underscoring the continued need for adaptive and inclusive global coordination mechanisms.
Paper Type: Research Articl
The relationship between family support and religiosity with gamophobia in career women: A correlational study
Full text linkExcessive fear of marital commitment, known as gamophobia, among adult career women has received limited scholarly attention, particularly within psychosocial and religiosity frameworks. This study investigated the effects of family support and religiosity on gamophobia among career women aged 33 years and older. A quantitative correlational approach was applied using multiple regression analysis. The study involved 47 career women recruited through purposive sampling. Data were gathered using standardized Likert type instruments and analyzed with SPSS version 26. The findings revealed that family support (β = –0.420, p < .001) and religiosity (β = –0.495, p < .001) were significant negative predictors of gamophobia, with the model explaining 52% of the variance (R² = 0.52). These results highlight the complementary roles of social support in fostering emotional regulation and religiosity in providing existential meaning, both of which contribute to reducing fear related to marital commitment. From a practical perspective, the findings suggest that intervention programs integrating family support enhancement and religiosity reinforcement may be effective in alleviating gamophobia among adult career women
Eating only the same food: How health concerns and gender influence food neophobia
Full text linkFood neophobia is an aversion to and avoidance of new foods, a phenomenon that can significantly influence dietary patterns and food choices throughout the lifespan. Food neophobia is thought to be influenced by personal factors, such as health concerns and gender. The purpose of this study was to determine the influence of health concerns and gender on food neophobia, with the hypothesis that both factors affect it. This study employed a cross-sectional design with participants aged 18–25 years. Data were collected using the Health Concern Scale and the Food Neophobia Scale, which were distributed via various social media platforms. A univariate general linear model was conducted to identify the influence of independent variables on food neophobia. The results indicate that health concerns positively influence food neophobia, whereas gender does not moderate the relationship between health concerns and food neophobia. These findings can serve as a basis for developing psychological interventions tailored to individuals with food neophobia based on their level of health concern
Self-Efficacy and Self-Regulated Learning among College Students in Digital Learning: A Meta-analytic Study
Full text linkThe success of students in digital learning depends largely on internal psychological factors such as self-efficacy and self-regulated learning (SRL). However, the strength and consistency of the relationship between these two constructs vary across studies. This meta-analysis synthesizes findings from seven empirical studies (total N = 1,971) published between 2020 and 2024 to estimate the magnitude of the relationship between self-efficacy and SRL among university students in digital learning contexts. Data were analyzed using the Sidik–Jonkman estimator model in Jamovi. Results revealed a large and significant overall effect size (r = 0.66, p < .001; 95% CI [0.29, 1.03]) with high heterogeneity (I² = 97.6%, Q = 221.73, p < .001). Publication bias analysis indicated no significant bias (Egger’s regression test p = .896; Fail-safe N = 1,861). This study extends previous correlational research by providing quantitative evidence on the robustness of the link between self-efficacy and SRL specifically in digital learning environments. The findings contribute to a deeper understanding of how students’ confidence in their learning abilities supports the development of effective self-regulation strategies, thereby informing future interventions to enhance digital learning outcomes in higher education
The Effect of Mental Health and Psychosocial Support Interventions on Depression, Anxiety, and Stress in Children with Physical Disabilities
Full text linkThis study investigated the effectiveness of Mental Health and Psychosocial Support (MHPSS) interventions in reducing depression, anxiety, and stress among children with physical disabilities. The intervention program integrated relaxation techniques, emotional literacy, and peer-based activities to promote adaptive coping and emotional resilience. A pretest–posttest design was employed, and statistical analysis indicated a significant improvement in psychological well-being, with large effect sizes across all variables. The greatest change occurred in depressive symptoms, followed by reductions in anxiety and stress levels. These findings suggest that structured and contextually adapted MHPSS programs effectively enhance emotional regulation and psychosocial adjustment. The study highlights the importance of school-based psychosocial initiatives in fostering inclusive education and reducing stigma toward children with disabilities
Analisis Performa Algoritma Smote-Bagging Dalam Klasifikasi Data Tidak Seimbang Dengan Metode Chi-Square Automatic Interaction Detection (CHAID)
Full text linkKlasifikasi data tidak seimbang sering menghadapi tantangan dalam mencapai keseimbangan antara sensitivitas dan spesifisitas. Penelitian ini menganalisis performa algoritma SMOTE-Bagging pada klasifikasi data tidak seimbang menggunakan metode Chi-Square Automatic Interaction Detection (CHAID), dengan studi kasus stunting pada balita tahun 2022 di Bojongsoang. SMOTE (Synthetic Minority Over-sampling Technique) digunakan untuk meningkatkan representasi kelas minoritas dalam dataset, kemudian digabungkan dengan teknik Bagging untuk meningkatkan kinerja klasifikasi. Hasil penelitian menunjukkan bahwa algoritma SMOTE-Bagging CHAID meningkatkan performa dalam klasifikasi data tidak seimbang, dengan peningkatan sensitivitas sebesar 65%, Area Under Curve (AUC) sebesar 42%, dan keseimbangan antara sensitivitas dan spesifisitas (G-Mean) sebesar 71%. Implementasi SMOTE-Bagging meningkatkan sensitivitas dan memberikan keseimbangan yang lebih baik antara sensitivitas dan spesifisitas
Optimasi Metode CART Menggunakan Metode Bagging Pada Studi Kasus Data Imbalance Berbasis Metode Adasyn
Full text linkPenelitian ini membahas mengenai permasalahan data imbalance yang menyebabkan kinerja dari model klasifikasi menjadi tidak optimal. Dalam penelitian ini menerapkan metode Adaptive Synthetic Sampling (ADASYN) untuk menangani permasalahan data imbalance, metode Classification and Regression Tree (CART) diterapkan sebagai metode klasifikasi pada dataset penyakit stroke, dan metode Boostrap Agregating (Bagging) untuk mengoptimalkan metode Cart. Penelitian ini bertujuan untuk mengetahui cara kerja dan performa dari penerapan metode Adasyn, Cart, dan Bagging dengan membangun tiga model klasifikasi yaitu model Cart, model Cart Adasyn, dan model Cart Adasyn Bagging. Hasil penelitian menunjukan model Cart menghasilkan nilai akurasi sebesar 94%, G-mean sebesar 0%, dan AUC sebesar 50%. Model Cart Adasyn menghasilkan nilai akurasi 78%, G-Mean 74% dan AUC 74%. Model Cart Adasyn Bagging menghasilkan nilai akurasi 78%, G-mean 76%, dan AUC 76%. Oleh karena itu, dapat disimpulkan bahwa kombinasi metode Cart, Adasyn, dan Bagging memberikan performa terbaik dalam mengatasi data tidak seimbang untuk klasifikasi penyakit stroke. Model Cart Adasyn Bagging terbukti lebih baik dalam memprediksi kedua kelas mayoritas dan kelas minoritas