Abstract:Penggunaan dompet digital yang terus meningkat menghasilkan banyak ulasan pengguna yang dapat dimanfaatkan untuk mengevaluasi kualitas layanan. Penelitian ini bertujuan meningkatkan akurasi klasifikasi sentimen pengguna…
dompet digital menggunakan metode Stacking Ensemble Machine Learning yang mengombinasikan Support Vector Machine (SVM), Multinomial Naïve Bayes (MNB), dan AdaBoost dengan Logistic Regression sebagai meta-learner. Data ulasan diproses melalui tahapan text preprocessing meliputi case folding, cleaning, tokenizing, stopword removal, stemming, dan pembobotan fitur menggunakan TF-IDF. Penyeimbangan data dilakukan dengan SMOTE, sedangkan evaluasi model menggunakan 5-Fold Cross-Validation. Hasil penelitian menunjukkan bahwa model Stacking Ensemble memperoleh akurasi rata-rata 80,55%, lebih tinggi dibandingkan algoritma dasar. Evaluasi menggunakan Confusion Matrix, Classification Report, dan ROC Curve juga menunjukkan peningkatan nilai precision, recall, F1-score, dan kemampuan diskriminasi model. Hasil ini menunjukkan bahwa pendekatan Stacking Ensemble Machine Learning efektif untuk meningkatkan akurasi klasifikasi sentimen pengguna dompet digital serta mendukung evaluasi kualitas layanan berbasis opini pengguna.
The rapid growth of digital wallet usage has generated a large volume of user reviews that can be utilized to evaluate service quality. This study aims to improve the accuracy of digital wallet user sentiment classification using a Stacking Ensemble Machine Learning approach that combines Support Vector Machine (SVM), Multinomial Naïve Bayes (MNB), and AdaBoost with Logistic Regression as the meta-learner. User reviews were processed through text preprocessing stages, including case folding, text cleaning, tokenization, stopword removal, stemming, and TF-IDF feature weighting. Synthetic Minority Over-sampling Technique (SMOTE) was employed to address class imbalance, while model performance was evaluated using 5-Fold Cross-Validation. The experimental results show that the proposed Stacking Ensemble model achieved an average accuracy of 80.55%, outperforming the individual base learners. Furthermore, evaluations based on the Confusion Matrix, Classification Report, and Receiver Operating Characteristic (ROC) Curve demonstrated improvements in precision, recall, F1-score, and the model's discriminative capability. These findings indicate that the proposed Stacking Ensemble Machine Learning approach is effective in improving the accuracy of digital wallet user sentiment classification and can serve as a reliable tool for supporting service quality evaluation based on user opinions.
Abstract:Sports-related minor injuries are common occurrences in school-based physical education activities, yet literacy regarding appropriate first aid management remains limited among teachers and students. This community service…
ice program aimed to improve knowledge, skills, and preparedness in managing minor sports injuries through a digital-based educational intervention entitled Smart Recovery. The program was implemented among junior high school (SMP) teachers and students in Banyumas Regency, Indonesia, using a participatory and technology-integrated approach. Activities consisted of program socialization, theoretical and practical training on minor injury management based on the RICE principle (Rest, Ice, Compression, Elevation), and the pilot implementation of the Smart Recovery digital application. Program effectiveness was evaluated using pre-test and post-test assessments, practical skill observations, and user feedback questionnaires. The results demonstrated a substantial improvement in participants’ knowledge, with average scores increasing from 62.4 to 88.6, and 95% of participants successfully performing the RICE procedures correctly after training. Furthermore, 92% of users reported that the digital application facilitated their understanding of injury management, while 87% perceived the platform as user-friendly and applicable for physical education learning. The integration of hands-on training and digital learning media proved effective in enhancing sports injury literacy, digital competence, and safety awareness in school settings. These findings indicate that the Smart Recovery program has strong potential as a sustainable and scalable model for promoting sports safety education within junior high schools.
Keywords : minor sports injury; RICE method; digital learning; Smart Recovery; physical education
Abstract:Pelatihan desain UI/UX (User Interface/User Experience) di Pondok Pesantren Terpadu Al Mumtaz bertujuan untuk meningkatkan keterampilan peserta, terutama santri dan mahasiswa Universitas Nahdlatul Ulama Yogyakarta, dalam…
merancang antarmuka pengguna yang efektif dan responsif. Workshop ini diselenggarakan sebagai bagian dari program pengabdian masyarakat untuk mendukung pengembangan website dan sistem informasi berbasis digital di lingkungan pesantren. Kegiatan pelatihan ini terdiri dari pengenalan konsep dasar UI/UX, praktik pembuatan wireframe, dan evaluasi desain aplikasi yang relevan dengan kebutuhan pesantren. Hasil pelatihan menunjukkan peningkatan pemahaman peserta terhadap prinsip-prinsip desain antarmuka pengguna, meskipun tantangan utama yang dihadapi adalah dalam menerjemahkan teori ke dalam desain yang lebih kompleks dan interaktif. Evaluasi terhadap desain yang dihasilkan menunjukkan perlunya pengembangan lebih lanjut dalam hal interaktivitas dan fungsionalitas untuk meningkatkan pengalaman pengguna. Sebagai tindak lanjut, disarankan untuk melakukan pelatihan lanjutan guna memperdalam keterampilan peserta dalam merancang aplikasi digital yang lebih canggih dan responsif. Workshop ini memberikan kontribusi positif terhadap digitalisasi Pondok Pesantren Al Mumtaz dan diharapkan dapat terus berlanjut dalam meningkatkan kualitas pengelolaan dan layanan pesantren berbasis teknologi.
Abstract:Abstract This study aims to analyze the implementation of the Integrated Licensing Service Information System (SIPINTER) at the Investment and One-Stop Integrated Services Office (DPMPTSP) of Tangerang Regency. The research…
earch is grounded in issues of low institutional responsiveness in following up on complaints, mismatches between actual service completion time and the Standard Operating Procedure (SOP), and uneven public socialization of the system. This study employs a descriptive method with a qualitative approach, using 3-dimensional e-government success model, namely Support, Capacity, and Value, as the analytical framework. Data were collected through field observation, regulatory documentation review, and in-depth interviews. The findings indicate that the Support dimension is backed by strong political will, allocated funding, and multi-channel socialization that has driven growth in registered users, though technical assistance responsiveness outside office hours remains weak. The Capacity dimension shows adequate human resources and staff competence, but the technological infrastructure remains prone to disruptions such as OTP delivery failures and server downtime. The Value dimension has produced tangible benefits in efficiency and service transparency, yet this value is undermined by document processing delays exceeding SOP standards, driven by a "digital red tape" practice among internal verification officers. This study recommends strengthening online technical support and adopting contingency resource mechanisms to improve service.
Keywords: SIPINTER, e-government, licensing service, DPMPTSP, Support-Capacity-Value model
Abstrak
Penelitian ini bertujuan untuk menganalisis Penerapan Sistem Informasi Pelayanan Perizinan Terpadu (SIPINTER) di Dinas Penanaman Modal dan Pelayanan Terpadu Satu Pintu (DPMPTSP) Kabupaten Tangerang. Latar belakang penelitian ini didasari oleh masih rendahnya responsivitas instansi dalam menindaklanjuti laporan, ketidaksesuaian durasi penyelesaian layanan dengan Standar Operasional Prosedur (SOP), serta sosialisasi yang belum merata kepada masyarakat. Penelitian menggunakan metode deskriptif dengan pendekatan kualitatif, dengan analisis berlandaskan model keberhasilan e-government tiga dimensi yaitu Support, Capacity, dan Value. Hasil penelitian menunjukkan bahwa dimensi Support telah didukung oleh political will yang kuat, alokasi anggaran, dan sosialisasi multikanal yang mendorong pertumbuhan pengguna terdaftar, namun masih lemah dalam responsivitas bantuan teknis di luar jam kerja. Dimensi Capacity menunjukkan kecukupan sumber daya manusia dan kompetensi petugas, tetapi infrastruktur masih rentan mengalami gangguan teknis seperti kegagalan pengiriman OTP dan server down. Dimensi Value menghasilkan manfaat nyata berupa efisiensi dan transparansi pelayanan, namun nilainya terdegradasi akibat keterlambatan penyelesaian dokumen yang melampaui SOP, yang dipicu oleh praktik birokrasi digital (digital red tape) di tingkat verifikator internal. Penelitian ini merekomendasikan penguatan dukungan teknis daring serta penerapan mekanisme kontingensi sumber daya untuk meningkatkan layanan.
Kata Kunci: SIPINTER, e-government, pelayanan perizinan, DPMPTSP, model Support-Capacity-Value
Abstract:Digitalization in education continues to increase, particularly in the use of learning media. However, various challenges associated with the emergence of eras based on birth years often lead to differing perspectives on…
improving the quality of education today. The process of integrating Powtoon learning media with a sense of responsibility in Gen Z students requires a comprehensive effort, considering that the assessment indicators for this attitude are crucial for learners in the modern era. Gen Z students (born 1997-2012) are digital natives who are pragmatic, creative, and highly dependent on technology for learning and personal branding. They favor visual-interactive methods, care about mental health/the environment, and focus on practical career outcomes (side hustles), but are prone to distraction and anxiety. Meanwhile, types of student attitude assessments generally include direct observation by lecturers, self-assessment, peer assessment, and daily journal entries. Aspects assessed include discipline, responsibility, honesty, cooperation, politeness, and initiative. This technique often uses a Likert scale or descriptive rubric. Therefore, it is crucial to pay attention to these items in the learning process at university. Powtoon is a web-based online animation and presentation creation platform that allows users to create engaging visual content, such as explainer videos, without requiring advanced graphic design skills. This application uses drag-and-drop features, ready-to-use templates, animated characters, and is supported by AI-based tools. This research uses a descriptive method, namely a research approach that aims to describe, explain, or map the characteristics of a phenomenon, situation, or population systematically, factually, and accurately based on data collected during the study. The results show the successful integration of Powtoon learning media with the responsible attitude of Gen Z students, evidenced by the existence of an orderly, understandable learning process and a high quality of understanding.
Abstract:PayLater services are one of the rapidly growing digital financial innovations widely utilised in fintech apps in Indonesia, including Kredivo and Akulaku. User reviews on the Google Play Store reflect a range of experiences,…
nces, from satisfaction with the ease of use of the service to complaints regarding bills, interest rates, late payment fees, credit limits, and app performance. This study aims to classify the sentiment of user reviews regarding PayLater services on the Kredivo and Akulaku apps using the Multinomial Naïve Bayes algorithm. Data was collected via web scraping from the Google Play Store and automatically labelled based on user ratings, with ratings of 1-2 classified as negative sentiment and ratings of 4-5 as positive sentiment, whilst a rating of 3 was excluded as it was considered ambiguous. Following a preprocessing stage comprising cleaning, case folding, tokenisation, stopword removal, and stemming, as well as feature extraction using TF-IDF, 3,652 reviews were obtained with a training-to-test data split ratio of 80:20. The results indicate that positive sentiment dominates the dataset at 56.49%, whilst negative sentiment accounts for 43.51%. Analysis by application revealed that Kredivo was dominated by positive sentiment (68.20%), whilst Akulaku was dominated by negative sentiment (51.70%). The Naïve Bayes multinomial model achieved an accuracy of 84.13%, with average precision, recall, and F1-score values of 0.84, demonstrating good and balanced classification performance across both sentiment classes.
Abstract:The rapid advancement of digital technology has fundamentally reshaped consumer behavior globally, with social media becoming the primary channel for product information seeking. Approximately 51% of Indonesians rely on…
user reviews before purchasing, and 62% have bought products under the direct influence of Social Media Influencers (SMIs). Despite this, the psychological mechanisms linking influencer characteristics to purchase intention remain fragmented across sectors and platforms. This systematic literature review synthesizes nineteen empirical studies published between 2020 and 2025, spanning beauty, fashion, gadget, and sustainable product sectors across multiple countries, to examine how Source Credibility (attractiveness, trustworthiness, expertise) and Parasocial Interaction (PSI) jointly shape consumer purchase intentions. Three core findings emerge: (1) trustworthiness is the sole credibility dimension that is universally and consistently significant across all contexts, fundamentally challenging assumptions about the equal contribution of the three credibility dimensions; (2) PSI operates not as a single mediator but as a multi-pathway psychological bridge through hedonic value, brand attitude, vicarious experience, and social-personal value whose specific active pathway is contingent on platform, product type, and cultural context; and (3) generational and cultural moderators systematically shape which credibility dimension most powerfully drives PSI. A Proposed Integrative Conceptual Framework and a synthetic cross-study comparison matrix (Table 2) consolidate these findings. Results offer strategic implications for influencer marketing practice and identify priority directions for future research.
Abstract:Digital transformation in healthcare services has significantly changed how healthcare professionals deliver clinical care, particularly in technology-based tertiary hospitals. Digital Health Competence (DHC) has become…
a crucial factor determining service effectiveness, clinical decision quality, and operational efficiency in modern healthcare organizations. This study aims to analyze the influence of digital health competence on healthcare service performance at King Abdullah Medical City (KAMC), Saudi Arabia, with work engagement as a mediating variable and organizational support as a moderating factor. This study employed a quantitative cross-sectional design. The sample consisted of 312 healthcare professionals at KAMC selected through stratified random sampling. Data were collected using a Likert-scale questionnaire ranging from 1 to 5 and analyzed using Structural Equation Modeling–Partial Least Squares (SEM-PLS). The findings revealed that digital health competence had a positive and significant effect on healthcare service performance (β = 0.412; p < 0.001). Digital competence also significantly influenced work engagement (β = 0.528; p < 0.001), while work engagement mediated the relationship between digital competence and healthcare service performance (β = 0.216; p < 0.001). Furthermore, organizational support strengthened the relationship between digital competence and work engagement. These findings indicate that digital competence not only improves service efficiency and quality but also functions as a psychological resource that enhances healthcare professionals’ motivation and dedication. Practically, the study provides important implications for hospitals in improving healthcare professionals’ digital competence through continuous training, the development of user-friendly digital systems, and adaptive organizational support.
Abstract:The increasing environmental issues, particularly the problem of plastic waste in Indonesia, as well as the gap between awareness and purchasing behavior of environmentally friendly products, have become important phenomena…
ena in this study. This study aims to analyze the effect of green promotion and green packaging on green purchase behavior, as well as to examine the role of environmental knowledge as a mediating variable in Easy Green Unilever products. In this study, data were collected through a survey distributed to users of Easy Green Unilever products in Indonesia, with a total of 400 respondents obtained. The research method employed a descriptive and quantitative approach, using non-probability sampling with a purposive sampling technique. In data processing, this study utilized SmartPLS software. The results indicate that green promotion and green packaging have a positive and significant effect on green purchase behavior. In addition, environmental knowledge is proven to have a significant effect on green purchase behavior and is able to mediate the relationship between green promotion and green packaging on green purchase behavior. The conclusion of this study emphasizes that increasing consumers’ environmental knowledge is a key factor in strengthening the effectiveness of green marketing strategies in encouraging environmentally friendly purchasing behavior.
Abstract:The growing emphasis on sustainable urban mobility has positioned electric vehicles (EVs) as a strategic component of transport transformation, particularly in developing cities with constrained infrastructure systems. One…
ne of the key challenges in supporting EV adoption is the provision of accessible charging infrastructure, which is closely related to user perception of acceptable service distance. This study aims to determine the optimal service distance of Electric Vehicle Charging Stations (EVCS) based on user preferences in Pekanbaru City, Indonesia, and to explore its implications for spatial planning. A quantitative approach was employed using a questionnaire survey of 51 respondents distributed across 15 administrative districts. All respondents were non-EV users, representing potential adopters in the early stage of transition. The survey captured preferred EVCS distance using interval-based categories, along with respondents’ awareness and intention toward EV adoption. The results indicate that the dominant preferred distance falls within the 2–4 km range, with an estimated mean of 3.08 km and a standard deviation of 1.98 km, resulting in an effective range of approximately 1–5 km. Considering urban road network constraints and detour factors, this range is further adjusted to 3–6 km to reflect more realistic travel conditions. These findings suggest that a relatively dense and accessible charging network is required to reduce range anxiety and support EV adoption. From a spatial planning perspective, a 4 km service threshold is proposed as a practical reference for EVCS deployment, representing the central tendency of user preferences within the adjusted range. This study highlights the importance of integrating user-based spatial parameters into infrastructure planning and contributes to the literature by emphasizing the role of perceived accessibility in shaping EV infrastructure demand in developing urban contexts.