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
Optimizing Function-Level Source Code Classification Using Meta-Trained CodeBERT in Low-Resource Settings Abednego Dwi Septiadi; Muhamad Awiet Wiedanto Prasetyo; Geusan Edurais Aria Daffa
Journal of Applied Data Sciences Vol 6, No 3: September 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i3.902

Abstract

This study investigates the effectiveness of a meta-trained transformer-based model, CodeBERT, for classifying source code functions in environments with limited labeled data. The primary objective is to improve the accuracy and generalizability of function-level code classification using few-shot learning, a strategy where the model learns from only a few labeled examples per category. We introduce a meta-learning framework designed to enable CodeBERT to adapt to new function types with minimal supervision, addressing a common limitation in traditional code classification methods that require extensive labeled datasets and manual feature engineering. The methodology involves episodic few-shot classification, where each episode simulates a low-resource task using five labeled and five unlabeled samples per function class. A balanced subset of Python functions was sampled from the CodeXGLUE benchmark, consisting of ten function categories with equal representation. The source code was preprocessed by removing comments and docstrings, then tokenized into a fixed length of 128 tokens to fit the model input format. The meta-trained CodeBERT was evaluated across 10 episodes, each representing a different task composition. Results show that the model achieves an average classification accuracy of 73.0%, with high accuracy on function categories characterized by unique syntax patterns, and lower performance on categories with overlapping logic or naming structures. Despite this variability, the model-maintained accuracy above 60% in all episodes. These findings suggest that meta-learning significantly enhances the adaptability of CodeBERT to unseen tasks under data-constrained conditions. This research demonstrates that meta-trained transformer models can serve as practical tools for real-time code analysis, particularly in integrated development environments and continuous integration pipelines. Future work may include extending the framework to other programming languages and incorporating semantic code representations to further reduce classification ambiguity.
Water Quality Prediction using Random Forest Algorithm and Optimization Deshinta Arrova Dewi; Aik Sam Wei; Leong Chi Lin; Chang Ding Heng
Journal of Applied Data Sciences Vol 5, No 3: SEPTEMBER 2024
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v5i3.348

Abstract

In the field of environmental conservation, the integration of Artificial Intelligence (AI) into pollution control strategies offers a transformative approach with significant potential. This paper presents a study on the application of AI techniques, specifically Random Forest algorithms, to predict and manage water quality in river systems. The objective of this research was to evaluate the performance of Random Forest models in comparison to Artificial Neural Networks (ANNs) for predicting the Water Quality Index (WQI). The study's findings revealed that the Random Forest model achieved a Mean Absolute Error (MAE) of 7.87 and a Root Mean Squared Error (RMSE) of 27.99, significantly outperforming the ANN model, which had a MAE of 121.40 and an RMSE of 215.04. These results demonstrate the superior accuracy and reliability of the Random Forest algorithm in capturing complex environmental data patterns. The novelty of this research lies in its comprehensive comparison of AI models for environmental monitoring, providing a data-driven approach to improving water quality management. This contribution is particularly relevant in the context of achieving Sustainable Development Goal (SDG) 6, which focuses on ensuring clean water and sanitation. By advancing traditional environmental planning methods with AI, this study highlights the potential of these technologies to make a substantial impact on environmental protection efforts.
A Hybrid YOLO–CNN Model for Automatic Detection and Severity Assessment of Atopic Dermatitis in Infant Images Debi Setiawan; Ramalia Noratama Putri; Sara Herlina; Achmad Nizar Hidayanto; Yuda Irawan; Naohiro Hohashi
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.v7i2.1212

Abstract

Atopic dermatitis is one of the most common skin diseases affecting infants and children worldwide and has a particularly high prevalence in tropical countries. Traditional diagnosis methods, which still rely on physical examinations and laboratory tests often face challenges such as delays, high costs, and limited facilities, thereby necessitating an artificial intelligence–based system that is more efficient and accurate. This study aims to develop a hybrid YOLO–CNN model for the automatic detection and severity classification of atopic dermatitis in infants. The dataset comprises 2,000 infant skin images, including lesions categorized as mild, moderate, and severe, obtained from an online repository and field observations conducted in three villages. The labeling process was performed by a specialist doctor to ensure clinical validity. In the first step, YOLO was used to detect the lesion area in real time by generating a bounding box. This produced a region of interest (ROI), which was subsequently analyzed by a CNN model employing transfer learning in the second step to determine the severity level. Experimental results indicate that YOLO achieved high detection performance, with an mAP@0.5 of 91.2% and an F1-score of 90.2%, while the CNN model attained an average accuracy of 85% and a macro-F1 score of 85% in classification. The visualization of predictions indicates that most lesions were detected with confidence levels ≥0.9, confirming the model’s consistency. These findings highlight the potential of the hybrid YOLO–CNN framework as a supportive system for digital clinical diagnosis, applicable to both mobile applications and teledermatology services, particularly in regions with limited medical personnel. Future research should employ larger, multi-center datasets and integrate explainable AI approaches to promote broader clinical adoption.
Nature-based Hyperparameter Tuning of a Multilayer Perceptron Algorithm in Task Classification: A Case Study on Fear of Failure in Entrepreneurship Theresia Ratih Dewi Saputri; Edwin Kurniawan; Caecilia Citra Lestari; Tony Antonio
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.v6i2.539

Abstract

Entrepreneurship plays a key role in generating economic growth, encouraging innovation, and creating job opportunities. Understanding which demographic, psychological, and socio-economic factors contribute to fear of failure in entrepreneurship is essential to developing proper standards in entrepreneurship education and policy. However, it remains challenging to accurately classify these factors, especially when balancing model performance with model complexity in a multilayer perceptron algorithm. An effective model requires the correct parameter setting via a hyperparameter tuning process. Adjusting each hyperparameter by hand requires significant effort and knowledge, as there are frequently multiple combinations to consider. Furthermore, manual tuning is prone to human error and may overlook optimal configurations, resulting in inferior model performance and prediction accuracy. This study evaluates nature-inspired optimization techniques, including particle swarm optimization (PSO), genetic algorithm (GA), and grey wolf optimization (GWO). Several parameters are tuned in the present multilayer perceptron model, including the number of hidden layers and the number of nodes in each hidden layer, learning rate, and activation functions. The used dataset which consists of 39 features from 333 samples captured individual fears, loss score, and computational efficiency as the required amount of time for finding the best parameter combination. Model accuracy performance scores are 45.16%, 53.76%, and 58.61% for GA, PSO, and GWO, respectively. Meanwhile their execution time are 10 minutes, 27 minutes, and 23 minutes, for GA, PSO, and GWO, respectively. Experiment results further reveal that each optimization algorithm has distinct advantages: GA excels at speedy convergence, PSO provides a robust exploration of hyperparameter space, and GWO offers remarkable adaptability to complicated parameter interdependencies. This study provides empirical evidence for the efficacy of nature-inspired hyperparameter modification in improving multilayer perceptron performance for fear of failure categorization tasks.
CO2 Emission Forecasting in Indonesia Until 2030: Evaluation of ETS Smoothing Prediction Models and Their Implications for Global Climate Change Mitigation Saepul Aripiyanto; Dewi Khairani; Ambran Hartono
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.v5i1.154

Abstract

The objective of this study is to predict CO2 emissions in Indonesia until 2030 utilizing the ETS smoothing prediction model in line with the pressing demand for viable climate change mitigation approaches. Through an assessment of the model's efficacy, several fundamental evaluation metrics have been identified. The research findings reveal that the Mean Absolute Error (MAE) stands at 146,154.40, presenting an overview of the average absolute disparity between the projected and actual CO2 emission values. The Mean Squared Error (MSE) of 21,838,251,772.37 characterizes the mean of the squared variances between projections and actual values, gauging the variability of predictive errors. The Root Mean Squared Error (RMSE) at 147,777.71, derived from the square root of MSE, reflects the degree of uncertainty in CO2 emission predictions. Simultaneously, the Mean Absolute Percentage Error (MAPE) of 7.24% provides an overview of the average percentage of absolute discrepancies between projections and actual values. Projections suggest that CO2 emissions could potentially reach 1 million tons in 2030. This evaluation furnishes an in-depth comprehension of the precision of the ETS smoothing model in the context of substantial emission escalation. The implications on the challenges of climate change mitigation become increasingly crucial, underscoring the immediacy of preemptive measures and sustainable policies. While the model delineates emission trends, it is imperative to acknowledge that these forecasts are subject to various influences, such as policy and technological shifts. Consequently, this study underscores the necessity for heightened awareness and the formulation of more efficacious policies to address the potential surge in CO2 emissions in the forthcoming years.
Modernizing Medicinal Plant Recognition: A Deep Learning Perspective with Data Augmentation and Hybrid Learning Preeti Nagrath; Malathy Batumalay; Dhruv Saluja; Harsh Kukreja; Devanshi Tegwal; Akash Saini
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.v6i4.795

Abstract

This study proposes a deep learning-based solution to address the longstanding challenge of accurately identifying Indian medicinal plants, which are vital to Ayurvedic pharmaceutics but often misidentified due to their morphological similarities. The objective is to develop a reliable, automated classification system using image processing and advanced neural network architectures. A dataset of 5,945 images representing 40 distinct medicinal plant species was sourced from Kaggle and augmented to 11,890 images using techniques such as flipping, rotation, and scaling to enhance diversity. The models tested include a baseline Convolutional Neural Network (CNN), transfer learning with DenseNet121, DenseNet169, and DenseNet201, a voting ensemble of these DenseNet variants, and a hybrid DenseNet201-LSTM architecture. Experimental results show that the CNN model achieved the lowest accuracy at 69.58%, while the hybrid DenseNet201-LSTM model reached the highest validation accuracy of 93.38%, with a precision of 94.74%, recall of 93.38%, and F1-score of 93.42%. These findings confirm the hybrid model’s superior ability to capture spatial and sequential dependencies in leaf features. The novelty of this work lies in the integration of DenseNet201 with LSTM for medicinal plant classification, which has not been widely explored in this domain. The study also acknowledges dataset scalability as a limitation and proposes future work involving dataset expansion through botanical collaborations, integration of environmental metadata, and deployment of a mobile application using TensorFlow Lite for real-time, low-resource implementation. Overall, the research contributes a robust and scalable framework for medicinal plant identification, promoting trust in traditional medicine, supporting conservation efforts, and enabling practical field-level applications in both rural and clinical settings.
Utilizing Structural Equation Modelling to Evaluate Factors Affecting Investment Capital Attraction and Sustainable Development in Vietnam Do Thi Lan Dai; Phan Thanh Tam
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.v5i4.428

Abstract

Vietnam and several nations globally are facing unparalleled effects of climate change, the progressive exhaustion of natural resources, and the global Covid-19 pandemic alongside non-traditional security threats. These tremendous challenges and occurrences underscore the significance of harmonious and balanced development among the three pillars of economics, society, and the environment. Sustainable development has emerged as an imperative necessity and an unavoidable trajectory in the evolution of human society. This study examines the principal aspects influencing investment capital attraction and sustainable development, consequently offering suggestions to enhance this process. The study employed two methodologies: qualitative research, executed via interviews, and concentrated on 15 economic expert group talks to modify the substance of observable factors to align with the business's features. Quantitative research was conducted on 800 representative managers from three Vietnamese provinces and one city to evaluate the model and research assumptions. The results indicate five elements influencing the attraction of investment capital, with a significance level of five percent, and the attraction of investment capital impacting sustainable development in Vietnam. This contribution enhances academic significance and is a reference for future studies on sustainable development in Vietnam. Five policy implications and contributions exist to advancing sustainable development in Vietnam, fostering innovation and enthusiasm. The novelty of this study is that sustainable development will establish Vietnam's working and living environment by concurrently advancing three dimensions: sustainable economic growth, a prosperous and equitable society, stable cultural diversity, a pristine environment, and preserved resources maintained sustainably. Consequently, a comprehensive framework of ethical principles for sustainable development encompasses the concepts of sustainability across the economic, social, and environmental dimensions.
Nonintrusive Arrhythmia Detection from Wrist Pulse Using NTSC Color Model in Eulerian Video Magnification Baby Lolita Basyah; Hustinawaty Hustinawaty; Miftahul Jannah
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.v7i2.1293

Abstract

Arrhythmia is a cardiovascular condition characterized by abnormal heart rhythms, such as tachycardia and bradycardia, which may lead to serious health complications if not detected early. This study proposes a non-invasive approach for screening tachycardia by extracting pulse signals from wrist video recordings using Eulerian Video Magnification (EVM) combined with the NTSC color space model. Subtle variations in skin color caused by blood flow, which are typically imperceptible to the human eye, are amplified using the EVM technique to enhance pulse-related motion signals. The NTSC color model is employed to separate luminance and chrominance components (YIQ), allowing more effective identification of pulse-induced color variations in the wrist region. The recorded wrist videos are processed through several stages, including spatial decomposition, temporal filtering, motion magnification, and pixel intensity extraction from the region of interest to obtain a temporal pulse signal. Peak detection is then applied to estimate heart rate in beats per minute (BPM). The performance of the proposed method is evaluated by comparing the estimated BPM values with reference measurements obtained from a Xiaomi Mi Band 2 wearable device. Experimental results based on 20 wrist video recordings demonstrate that the proposed method achieves approximately 96% agreement between the estimated BPM values and the reference measurements. Quantitative evaluation using Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and correlation analysis further confirms the consistency of the proposed approach. These results indicate that the integration of Eulerian Video Magnification with the NTSC color model has potential as a low-cost and non-contact method for preliminary tachycardia screening and remote cardiovascular monitoring.
Data mining for Education Sector, a proposed concept Ammar Salamh Mujali Al-Rawahnaa; Anas Yahya Bader Al Hadid
Journal of Applied Data Sciences Vol 1, No 1: SEPTEMBER 2020
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v1i1.6

Abstract

Data mining is very much needed in various fields In accessing a large amount of data requires time and a high level of accuracy. In higher education the potential influence of data mining on the learning processes and outcomes of the students was realized. Especially in the field of education, knowing almost every educational institute, both public and private, has thousands of data from students with a variety of different programs and subjects. Understanding the benefits of data retrieval will facilitate the course of education itself. The use of Data mining in education will be useful in developing a student-focused strategy and in providing the correct tools that institutions would be able to use for quality improvement purposes. In this paper, we will find out the benefits of applying data mining in the education sector using classification, prediction, association and clustering methods.
Dynamic Replica Management Strategy Based-on Data Accessing Popularity for Load Balancing and Optimizing Network Performance in Cloud Storage S. Thavamani; Siti Sarah Maidin; S. T. Varun
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.v6i2.674

Abstract

Network performance plays a vital role in organizational efficiency, where large volumes of data, fast transmission, and low latency significantly enhance productivity and reduce downtime. Cloud storage offers a service model that enables remote data management and efficient content distribution. In such systems, data replication is widely used to improve availability, reliability, fault tolerance, and throughput. However, static replication policies often allocate replicas during system initialization, failing to adapt to the dynamic and heterogeneous nature of cloud environments. These environments are susceptible to challenges such as data loss, node failures, and fluctuating demand, which can degrade service quality. To address this, we propose a dynamic replica management strategy that considers data popularity, active peer participation, and peer capacity. Virtual peers are grouped into strong, medium, and weak clusters based on their weight values, which are derived from bandwidth, CPU speed, memory size, and access delay. Content is categorized into Class I, II, and III based on access frequency. Highly popular data (Class I) is replicated in strong clusters, while less frequently accessed data is placed in medium and weak clusters. A hierarchical routing mechanism ensures that queries are directed to the appropriate cluster. The proposed system was implemented and evaluated through simulations. Results show up to 25% improvement in throughput, 20% reduction in packet drops, 97% query efficiency, and decreased bandwidth utilization under high load. By maintaining optimal replica counts without compromising availability, the system supports cloud SLA compliance while minimizing overhead. This solution is aligned with the ninth UN Sustainable Development Goal: Industry, Innovation, and Infrastructure.