cover
Contact Name
Husni Teja Sukmana
Contact Email
husni@bright-journal.org
Phone
+62895422720524
Journal Mail Official
jads@bright-journal.org
Editorial Address
Gedung FST UIN Jakarta, Jl. Lkr. Kampus UIN, Cemp. Putih, Kec. Ciputat Tim., Kota Tangerang Selatan, Banten 15412
Location
Kota adm. jakarta pusat,
Dki jakarta
INDONESIA
Journal of Applied Data Sciences
Published by Bright Publisher
ISSN : -     EISSN : 27236471     DOI : doi.org/10.47738/jads
One of the current hot topics in science is data: how can datasets be used in scientific and scholarly research in a more reliable, citable and accountable way? Data is of paramount importance to scientific progress, yet most research data remains private. Enhancing the transparency of the processes applied to collect, treat and analyze data will help to render scientific research results reproducible and thus more accountable. The datasets itself should also be accessible to other researchers, so that research publications, dataset descriptions, and the actual datasets can be linked. The journal Data provides a forum to publish methodical papers on processes applied to data collection, treatment and analysis, as well as for data descriptors publishing descriptions of a linked dataset.
Articles 1,270 Documents
Optimization of Recommender Systems for Image-Based Website Themes Using Transfer Learning Arif Mu'amar Wahid; Taqwa Hariguna; Giat Karyono
Journal of Applied Data Sciences Vol 6, No 2: MAY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i2.671

Abstract

Recommender systems play a crucial role in personalizing user experiences in e-commerce, digital media, and web design. However, traditional methods such as Collaborative Filtering and Content-Based Filtering struggle to account for visual preferences, limiting their effectiveness in domains were aesthetics influence decision-making, such as website theme recommendations. These systems face challenges such as data sparsity, cold-start problems, and an inability to capture intricate visual features. To address these limitations, this study integrates Convolutional Neural Networks (CNNs) with advanced recommendation models, including Inception V3, DeepStyle, and Visual Neural Personalized Ranking (VNPR), to enhance the accuracy and personalization of visually-aware recommender systems. A quantitative research approach was employed, using controlled experiments to evaluate different combinations of feature extractors and recommendation models. Data was sourced from ThemeForest, a widely used platform for website themes, and underwent preprocessing to ensure consistency. The models were evaluated using precision, recall, F1 score, Mean Average Precision (MAP), and Normalized Discounted Cumulative Gain (NDCG) to measure recommendation quality. The results indicate that Inception V3 + VNPR outperforms other model combinations, achieving the highest accuracy in personalized theme recommendations. The integration of transfer learning further improved feature extraction and performance, even with limited training data. These findings underscore the importance of combining deep learning-based feature extraction with recommendation models to improve visually-driven recommendations. This study provides a comparative analysis of CNN-based recommender systems and contributes insights for optimizing recommendations in visually complex domains. Despite improvements, challenges such as dataset diversity remain a limitation, affecting generalizability. Future research could explore alternative CNN architectures, such as ResNet and DenseNet, and incorporate user feedback mechanisms to further enhance recommendation accuracy and adaptability.
Machine Learning Classifier Algorithms for Ransomware Lockbit Prediction Ibrahiem M. M. El Emary; Khalil A. Yaghi
Journal of Applied Data Sciences Vol 5, No 1: JANUARY 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i1.161

Abstract

Advanced virus known as ransomware has been spreading quickly in recent years, resulting in considerable financial losses for a variety of victims, including businesses, hospitals, and people. Modern host-based detection techniques need to first infect the host in order to spot abnormalities and find the malware. When the system is infected, it can already be too late because some of the assets have been exfiltrated or encrypted by the malware. On the other hand, as most ransomware families attempt to connect to command-and-control servers before to executing their damaging payloads, network-based methods can be helpful in detecting ransomware attacks. Therefore, one of the most important methods for early identification can be a detailed examination of ransomware network activity. This study presents a thorough behavioral analysis of the ransomware LockBit. In early 2022, ransomware, particularly targeting data on endpoints in Indonesia, was enough to horrify the news online. LockBit ransomware is one of the ransomwares that is particularly worrisome in Indonesia, so study is required to combat the ransomware. Static and dynamic analyses are used to study the ransomware; the former involves deciphering the portable executable (PE) file, while the latter involves actually running the ransomware. These analyses will reveal the impurity and resolve of the LockBit ransomware. Examine the running operations, the resources utilized, the network activities the ransomware performed, and the effect it had on the impacted operating system to try to build a scenario for preventative measures. The real effects of the ransomware-as-a-service (Raas) attacks conducted by the LockBit ransomware are demonstrated in this research. In this work, we describe an attribute selection-based system for identifying and avoiding ransomware that uses a variety of machine learning techniques, such as neural network-based frameworks, to classify the malware's security grade. We used a range of machine learning approaches, such as Decision Tree-DT, Random Forest-RF, Naive Bayes-NB, and Logical Regression-LR based classifiers, on a selected set of attributes for ransomware detection. The results of the study demonstrate that the Random-Forest predictor outperformed different classifiers by achieving the best accuracy, precision, recall, and F1-Score.
Comparative Analysis of Novel Deep Reinforcement Learning Methods for Food Distribution Optimization Jeperson Hutahaean; Yessica Siagian; Endra Saputra
Journal of Applied Data Sciences Vol 6, No 4: December 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i4.956

Abstract

Uneven food distribution across various regions in Indonesia often results in supply-demand imbalances, leading to price surges, stock shortages, and overall market instability. This challenge is compounded by the limitations of conventional distribution systems, which are ill-equipped to respond to rapidly changing market dynamics. In response, this study introduces a novel, AI-driven approach by implementing Deep Reinforcement Learning (DRL) to optimize food distribution policies using real-world data. Specifically, we perform a comparative evaluation of four emerging DRL models—Double Deep Q-Network (Double DQN), Dueling DQN, Proximal Policy Optimization (PPO), and Advantage Actor-Critic (A2C)—to determine their effectiveness in learning adaptive distribution strategies from national food logistics data provided by Indonesia’s Central Bureau of Statistics (BPS). Each model was trained within a custom simulation environment based on the Markov Decision Process (MDP) framework and evaluated using five core performance metrics: cumulative reward, average reward, success rate, sample efficiency, and best reward. The results reveal that A2C consistently outperformed the other models, delivering the highest average reward and most stable training performance, while PPO demonstrated strong efficiency and success rate. These findings underscore the potential of policy-gradient methods—particularly A2C—as robust and intelligent solutions for dynamic food logistics management. This research offers one of the first comparative benchmarks of DRL methods in the food distribution domain and highlights their applicability for future integration into national AI-powered logistics systems.
Novel Predictive Framework for Student Learning Styles Based on Felder-Silverman and Machine Learning Model Wiga Maulana Baihaqi; Rujianto Eko Saputro; Fandy Setyo Utomo; Sarmini Sarmini
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.408

Abstract

This study analyzes data from the Open University Learning Analytics Dataset to evaluate how students' interactions with Virtual Learning Environment (VLE) materials influence their final outcomes. This research aims to formulate and build a novel predictive framework based on the Felder-Silverman and Machine Learning Model for student learning styles. Based on these objectives, this research provides novelty and contributions since it enhances student data analysis, uses a learning model using Felder-Silverman Learning Style Model (FSLSM) to give a more comprehensive understanding of students' learning styles, and improves prediction accuracy by introducing Artificial Neural Network (ANN) and feature selection using Random Forest. The data used includes 3 main files: vle.csv, which contains information about the materials and activities in the VLE; studentVle.csv, which records students' interactions with the materials; and studentInfo.csv, which provides demographic information of students and their final outcomes. The analysis process involved data merging and processing, including handling of missing values, data type conversion, as well as mapping activity types to learning style features based on the FSLSM. We use the Random Forest feature selection method, as well as data imbalance handling techniques such as oversampling, to improve model performance. The applied classification models include Logistic Regression, K-Nearest Neighbor, Random Forest, Support Vector Machine (SVM), and ANN. The analysis results showed that after tuning, the Random Forest model achieved 97% accuracy, while SVM achieved 97% accuracy as well, with better performance than previous studies. This research highlights the importance of comprehensive data integration and appropriate processing techniques in improving the accuracy of student learning style prediction. Based on the increase in accuracy results, it can be beneficial for more effective personalized learning and improve our understanding of students' learning style preferences. The research advances knowledge and provides practical applications for educators to tailor their teaching strategies.
On the Use of Zero-Inflated Mixed Models for Count Data: A Simulation and Empirical Evidence Anang Kurnia; Zafira Fakhriyah; Kusman Sadik; Dian Handayani
Journal of Applied Data Sciences Vol 7, No 2: May 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i2.1303

Abstract

This paper evaluates the performance of classical count regression models (Poisson, Negative Binomial, Generalized Poisson), zero-inflated models (Zero-Inflated Poisson/ZIP, Zero-Inflated Negative Binomial/ZINB, Zero-Inflated Generalized Poisson/ZIGP), and zero-inflated mixed models (ZIPMM, ZINBMM, ZIGPMM) for over-dispersed count data, particularly due to excess zeros and unobserved heterogeneity. Using simulation and empirical studies, we evaluated the performance of the models based on their predictive capability and their ability to yield valid inferences through hypothesis testing. The simulation, replicated 1000 times, involves 27 scenarios that combine various sample sizes, proportions of zero counts, and response variable distributions. Our findings indicate that ZIGPMM and ZINBMM provide the smallest root mean square error (RMSE) values. Although the Poisson model yields a relatively small RMSE, it does not adequately account for overdispersion, leading to underestimated standard errors and potentially misleading significance tests. The negative binomial model yields dispersion estimates closest to 1, indicating good performance, whereas ZIGP, ZINB, ZIGPMM, and ZINBMM perform better when zero counts are extremely high. Empirical analysis of data on under-five mortality due to pneumonia in Java Island, Indonesia, indicates that ZINB, ZINBMM, and ZIGPMM have the smallest Akaike Information Criterion (AIC), making them the most suitable models. These models show that exclusive breastfeeding and vitamin A have no significant effect on under-five child mortality due to pneumonia, while severe malnutrition has a statistically significant impact (α=0.05).
Factors Influencing the Intention to Use Insurance Technology (Insurtech) Among Generation Z Using the Extended D-M Model M. Fankar Umran; Haris Maupa; Agustinus Purna Irawan; Andi Muhammad Sadat
Journal of Applied Data Sciences Vol 6, No 2: MAY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i2.726

Abstract

This study investigates the factors influencing Generation Z’s intention to use Insurtech in Indonesia using an extended DeLone and McLean model. The research introduces two additional variables: perceived trust and regulatory expectancy. Data were collected via an online survey of 431 Generation Z respondents aged 17 and above, residing in ten major Indonesian cities: Jakarta, Bandung, Semarang, Yogyakarta, Surabaya, Denpasar, Palembang, Medan, Balikpapan, and Makassar, all with a basic understanding of Insurtech. The questionnaire included demographic questions and research variables measured on a five-point Likert scale. Data were analyzed using Structural Equation Modeling (SEM) through Smart PLS 4. Descriptive analysis revealed that most respondents were aged 25-28 years, predominantly female, residing in Jakarta, employed in private sectors, with monthly expenditures below USD 300, and holding a bachelor’s degree. The analysis indicated that respondents viewed Insurtech positively, noting its organized information, flexible services, knowledgeable providers, honest services, and legal protection of personal data. Additionally, respondents expressed a strong interest in using Insurtech soon. The measurement model evaluation confirmed the validity and reliability of all indicators based on convergent validity, discriminant validity, and reliability tests. The structural model analysis showed that the independent variables explained 57% of the variance in intention to use Insurtech and 69% in perceived trust. Hypothesis testing revealed that information quality, system quality, service quality, and regulatory expectancy positively influenced both intentions to use Insurtech and perceived trust. However, contrary to expectations, perceived trust did not significantly affect the intention to use Insurtech. This finding suggests that for Generation Z, trust may be considered a baseline expectation, with factors like system and service quality playing a more direct role in their adoption decisions. Additionally, no significant mediation effects were found. The model demonstrated strong predictive relevance and good fit, confirmed by Q², NFI, and SRMR values.
Cutting-Edge AI Approaches with MAS for PdM in Industry 4.0: Challenges and Future Directions Shadia Yahya Baroud; Nor Adnan Yahaya; Abdelrafe M. Elzamly
Journal of Applied Data Sciences Vol 5, No 2: MAY 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i2.196

Abstract

Integrating Artificial Intelligence (AI) within Industry 4.0 has propelled the evolution of fault diagnosis and predictive maintenance (PdM) strategies, marking a significant shift towards smarter maintenance paradigms in the mechatronics sector. With the advent of Industry 4.0, mechatronic systems have become increasingly sophisticated, highlighting the critical need for advanced maintenance methodologies that are both efficient and effective. This paper delves into the confluence of cutting-edge AI techniques, including machine learning (ML) and deep learning (DL), with multi-agent systems (MAS) to enhance fault diagnosis precision and facilitate PdM in the context of Industry 4.0. Specifically, we explore the use of various ML models, including Support Vector Machines (SVMs) and Random Forests (RFs), and DL architectures like Convolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs), which have been effectively oriented to analyses complex industrial data. Initially, the study examines the progress in AI algorithms that accelerate fault identification by leveraging data from system operations, sensors, and historical trends. AI-enabled fault diagnosis rapidly detects irregularities and discerns the fundamental causes, thereby minimizing downtime and enhancing system reliability and efficiency. Furthermore, this paper underscores the adoption of AI-driven PdM approaches, emphasizing prognostics that predict the Remaining Useful Life (RUL) of machinery. This predictive capability allows for the strategic scheduling of maintenance activities, optimizing resource use, prolonging the lifespan of expensive assets, and refining the management of spare parts inventory. The tangible advantages of employing AI for fault diagnosis and PdM are showcased through a case study from authentic mechatronics implementations. This case study highlights successful implementations, documenting real-world challenges such as data integration issues and system interoperability, and elaborates on the strategies deployed to navigate these obstacles. The results demonstrate improved operational reliability and cost savings and shed light on the pragmatic considerations and solutions that facilitate the adoption of AI and MAS in industrial applications. The paper also navigates the challenges and prospective research avenues in applying AI within the mechatronics domain of Industry 4.0, setting the stage for ongoing innovation and exploration in this transformative domain.
Improving University Ranking Robustness Using Rank Geometric Weight Integration with CoCoSo Method for Reducing Ordinal Weighting Instability Septi Andryana; Teddy Mantoro; Achmad Benny Mutiara; Ernastuti Ernastuti; Prihandoko Prihandoko; Aris Gunaryati
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.v7i1.1024

Abstract

This study lies in the field of decision support systems, focusing on the application of Multi-Criteria Decision Making (MCDM) for ranking alternatives based on predefined organizational criteria. A persistent challenge in this domain is the instability and subjectivity of ordinal weighting methods - such as Rank Order Centroid (ROC), Rank Sum (RS), Rank Reciprocal (RR), and Rank Order Distribution (ROD), which derive weights solely from rank positions, often leading to inconsistent and unreliable outcomes. To address this, this study introduces Rank Geometric (RG) weights, a geometric mean aggregation of ROC, RS, RR, and ROD designed to reduce subjectivity, stabilize weight distribution, and enhance robustness. By using the Combined Compromise Solution (CoCoSo) method, the RG against Times Higher Education’s (THE) official weights were evaluated, and the four individual ordinal methods, applied to the top 10 Indonesian universities across five THE 2025 ranking criteria. Empirical results show that RG-CoCoSo produces stronger and more consistent correlations with THE’s rankings than THE-CoCoSo, as validated by Spearman and Pearson correlation tests, with a p-value of 0.0251. This study contributes a practical, data-driven weighting framework that strengthens the reliability of MCDM-based institutional performance evaluation and can be generalized to other ranking contexts.
Analyzing the Impact of Publicity and e-WOM on Indonesian Tourists’ Visit Intention to Seoul through Destination Awareness and Preference: A Structural Equation Modeling Approach Ghassani Herstanti; Usep Suhud; Agung Wahyu Handaru
Journal of Applied Data Sciences Vol 5, No 4: DECEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i4.531

Abstract

This research explores the role of digital publicity and electronic word of mouth (E-WOM) in shaping Indonesian tourists' intentions to visit Seoul. By integrating destination awareness and preference as mediating variables, the study provides a holistic view of how publicity and EWOM interact to influence visit intentions. Digital publicity, including news coverage and promotional content on social media, raises initial awareness of Seoul by highlighting its attractions, culture, and experiences. E-WOM, expressed through online reviews, travel blogs, and social media shares, further enhances the perception of Seoul by providing authentic, peer-driven narratives. These user-generated insights are particularly impactful, as they foster trust and add an emotional dimension to tourists’ perception of the destination. Using a structural equation modeling approach, the study analyzes survey responses from Indonesian tourists to validate six core hypotheses, examining the direct and indirect effects of publicity and E-WOM on destination awareness, preference, and visit intention. Results indicate that both digital publicity and E-WOM significantly contribute to tourists' awareness and preference for Seoul, with preference being a particularly strong predictor of visit intention. The findings underscore the importance of aligning digital publicity efforts with targeted E-WOM strategies, enabling tourism marketers to build both cognitive awareness and emotional appeal, which ultimately drive visit intention. These insights are valuable for tourism stakeholders aiming to enhance destination marketing strategies, as they suggest that a combined approach—leveraging both structured publicity and organic E-WOM—can effectively increase a destination’s appeal. By focusing on creating authentic, accessible content and fostering positive online word of mouth, tourism authorities can better attract tourists and establish Seoul as a top choice for Indonesian travelers.
Deep Deterministic Policy Gradient for Simulation-Based Control of Water Quality in Nano Aquatic Systems Iwan Fitrianto Rahmad; Syahril Efendi; Poltak Sihombing; Henny Febriana Harumy
Journal of Applied Data Sciences Vol 7, No 3: September 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v7i3.1447

Abstract

This study proposes a simulation-based Deep Deterministic Policy Gradient (DDPG) framework for water-quality control in nano aquatic systems. Nano tanks are highly sensitive to small disturbances because their limited water volume reduces buffering capacity and causes rapid changes in dissolved oxygen, ammonia, pH, temperature, biological oxygen demand, and chemical oxygen demand. To represent these coupled dynamics, a nonlinear simulation model adapted from the Continuously Stirred Tank Reactor concept is developed and implemented as an OpenAI Gym-compatible reinforcement learning environment. The DDPG agent learns continuous control actions related to aeration, feeding, and filtration through repeated interaction with the simulated nano-tank environment. The proposed nonlinear CSTR-DDPG framework is evaluated against a linear-model DDPG baseline using RMSE, cumulative reward, and closed-loop control performance. Simulation results show that the nonlinear model reduced RMSE by 45.2% for dissolved oxygen, 64.0% for ammonia, and 61.3% for pH compared with the linear baseline. The DDPG agent also achieved a 41.7% higher cumulative reward under the same reward structure. These findings indicate that nonlinear simulation can provide a more informative training environment for DDPG-based water-quality control. However, the present study remains limited to simulation-based evaluation, and physical validation using calibrated sensors, actuators, communication-delay analysis, and real nano-tank experiments is required in future work.

Page 76 of 127 | Total Record : 1270