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
An Effective Investigation of Genetic Disorder Disease Using Deep Learning Methodology B. Vidhya; B. L. Shivakumar; Siti Sarah Maidin; Jing Sun
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.v5i3.370

Abstract

This study evaluates the performance of four neural network models—Artificial Neural Network (ANN), ANN optimized with Artificial Bee Colony (ANN-ABC), Multilayer Feedforward Neural Network (MLFNN), and Forest Deep Neural Network (FDNN)—across different iteration levels to assess their effectiveness in predictive tasks. The evaluation metrics include accuracy, precision, Area Under the Curve (AUC) values, and error rates. Results indicate that FDNN consistently outperforms the other models, achieving the highest accuracy of 99%, precision of 98%, and AUC of 99 after 250 iterations, while maintaining the lowest error rate of 2.8%. MLFNN also shows strong performance, particularly at higher iterations, with notable improvements in accuracy and precision, but does not surpass FDNN. ANN-ABC offers some improvements over the standard ANN, yet falls short compared to FDNN and MLFNN. The standard ANN model, though improving with iterations, ranks lowest in all metrics. These findings highlight FDNN's robustness and reliability, making it the most effective model for high-precision predictive tasks, while MLFNN remains a strong alternative. The study underscores the importance of model selection based on performance metrics to achieve optimal predictive accuracy and reliability. 
Adaptive Integration of Optuna Optimization and Stacking Ensemble Learning for Automated Work Competency Classification Mutiana Pratiwi; Sarjon Defit; Muhammad Tajuddin
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.v7i2.1228

Abstract

Artificial intelligence and machine learning are increasingly used to automate analytical and decision processes, including the evaluation of human competencies. However, traditional models often face challenges in accuracy and generalization when applied to linguistic data from interviews. This study aims to develop a model that integrates Optuna optimization and stacking ensemble learning to enhance the accuracy and interpretability of competency classification. Interview transcript data were processed using natural language processing techniques such as cleaning, tokenization, case folding, stopword removal, and stemming to ensure textual consistency. The text was then transformed into numerical representations using term frequency inverse document frequency weighting. To handle class imbalance, the synthetic minority oversampling technique was employed. Optuna was applied to optimize the hyperparameters of base models, including support vector classifier, Naïve Bayes, random forest, gradient boosting, and XGBoost. These optimized models were combined through a stacking ensemble to form the final classifier. The proposed model achieved an accuracy of 94 percent and a precision of 95 percent with macro and weighted F1 scores of 0.94. The results demonstrate stable and balanced performance across all competency categories, including analytical thinking, initiating action, problem solving, and work standards. Comparative analysis with previous studies in sentiment analysis, medical diagnosis, and financial forecasting confirmed that the integration of Optuna and stacking produces more robust and generalizable outcomes. The integration of Optuna optimization and stacking ensemble learning effectively improves classification performance while maintaining interpretability. The model demonstrates strong potential for automated competency evaluation in recruitment and human resource analytics. This framework can be extended to other linguistic datasets to support transparent and data-driven decision-making in artificial intelligence applications.
Development of Skyline Query Algorithm for Individual Preference Recommendation in Streaming Data Ruhul Amin; Taufik Djatna; Annisa Annisa; Sukaesih Sitanggang
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.v6i2.599

Abstract

The ability of a recommendation system to deliver relevant outcomes is significantly influenced by its adaptability to the dynamic nature of individual user preferences. Data-streaming-based recommendation systems face substantial challenges in aligning recommendations with rapid shifts in user preferences. Previous research on the development of skyline query algorithms has predominantly focused on processing efficiency and parallel performance optimization yet has not addressed the dynamic nature of individual user preferences—an essential factor for generating relevant and responsive recommendations in streaming data environments. This study aims to develop a skyline query algorithm called Distributed Data Skyline (DDSky) to provide recommendations based on dynamic individual user preferences within data-streaming contexts. DDSky leverages the Recency, Frequency, Monetary, and Rating (RFMRT) model to capture real-time changes in user preferences. This model is integrated with parallel skyline computation and structured to enhance the data processing efficiency on a large scale. The parallel processing approach divides tasks into smaller subtasks executed simultaneously across multiple threads. This strategy enables the simultaneous processing of attributes such as price, distance, and individual user preferences, thereby delivering relevant and responsive recommendations to real-time changes in user preferences. The DDSky algorithm was evaluated using a local dataset from the JALITA application and compared with the Eager algorithm. The results demonstrated that DDSky outperformed Eager, achieving an average recall value of 0.45 and an F1-measure of 0.55, compared to Eager's recall value of 0.33 and F1-measure of 0.47. Furthermore, DDSky achieved an average precision of 0.73, which closely approached Eager's precision of 0.82. Additionally, DDSky exhibited optimal throughput performance for datasets containing up to 10,000 items with high flexibility across various data types. With its unique technical approach, DDSky delivers more responsive and relevant recommendations to dynamic user preferences, establishing its superiority in data-streaming-based recommendation systems.
Applying Structural Equation Modeling to Explore the Intention to Use Midi Kriing App Suwandi S. Sangadji; Tanti Handriana; Nugrahini Susantinah Wisnujati; Sarbaini A. Karim
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.v5i1.157

Abstract

In the rapidly evolving digital landscape, the surge in e-commerce transactions underscores the need for innovative strategies to enhance user satisfaction, trust, and sustainable app usage. This research focuses on the Midi Kriing App, operated by PT Midi Utama Indonesia Tbk, a key player in the e-commerce industry. The study aims to bridge knowledge gaps by investigating factors influencing user intention, specifically e-service quality and e-trust, and their impact on user satisfaction. Employing a quantitative approach with an associative design, data was gathered from 190 Midi Kriing App users in Surabaya, Indonesia. Structural Equation Modeling (SEM), particularly Partial Least Squares (PLS) in SmartPLS, was utilized to explore relationships between variables. Research findings indicate that e-service quality and e-trust significantly impact user satisfaction, with a p-value of 0.00. Similarly, user satisfaction significantly influences the intention to use the Midi Kriing App, with a p-value of 0.00. Among these hypotheses, the statistical t-value of user satisfaction with the intention to use the Midi Kriing App, at 9.871, is higher than the relationship between e-service quality and e-trust with user satisfaction. Nevertheless, these hypothesis tests confirm statistically significant relationships, supporting the reliability and significance of each construct's measurement instruments. In conclusion, this research emphasizes the pivotal role of satisfaction in its relation to e-service quality, e-trust, and the intention to use the Midi Kriing App. Managerial implications stress the importance of enhancing these factors to drive app usage. Improving e-service quality can be achieved through active efforts such as enhancing responsiveness, reliability, and user-friendliness. Similarly, building e-trust involves securing user data and providing a positive user experience.
A Data-Driven Training Kit to Enhance the Note Recording Skills of Music Learners Thaworada Chantanasut
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.v6i4.1063

Abstract

Music note recording is a fundamental skill in music education, yet many undergraduate learners struggle due to the lack of structured, self-guided practice resources. This study aimed to develop and evaluate a data-driven training kit designed to Enhance the Note Recording Skills of Music Learners. A total of 20 first-year students from a Bachelor of Education program in Western Music and Vocal Education were selected through purposive sampling. Research instruments included the training kit, expert evaluation forms, pre- and post-tests, student satisfaction surveys, and behavior observation checklists. Quantitative data were analyzed using mean, standard deviation, and paired sample t-tests. Results showed a statistically significant improvement in students' performance after the intervention (t = 18.789, p .001), with a large effect size (Cohen’s d = 2.53). Experts rated the kit highly (M = 4.86, SD = 0.29), and students reported very high satisfaction (M = 4.94, SD = 0.14). These findings support the kit’s effectiveness as an engaging and pedagogically sound tool for developing music note recording skills in higher education settings. 
Ensembling Methods for Data Privacy in Data Science N Mahendiran; B L Shivakumar; Siti Sarah Maidin; Hao Wu
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.v5i4.341

Abstract

The rapid advancement of technology has unified systems, data storage, applications, and operations, providing continuous services to organizations. However, this integration also introduces new vulnerabilities, particularly the risk of cyber-attacks. Malware and digital piracy pose significant threats to data security, with the potential to compromise sensitive information, leading to severe financial and reputational damage. This study aims to develop an effective method for detecting malware-infected files on storage devices within the Internet of Things (IoT) environment. The proposed approach utilizes a stacked regression ensemble for data pre-processing and the Sea Lion Optimization Algorithm (sLOA) to extract salient features, enhancing the classification process. Using malware data from an intrusion detection dataset, an ensemble classification technique is applied to identify malicious infections. The experimental results demonstrate that the proposed method achieved an accuracy of 98%, a precision of 99.6%, a recall of 96%, and an F-measure of 95% by the final iteration, significantly outperforming existing techniques in addressing cyber-security challenges within IoT systems.
Development of New Identification Formula to Extract Organic Fertilizer Content Based on Organic Fertilizer Image Agung Ramadhanu; Mardison Mardison; Halifia Hendri; Febri Hadi; Larissa Navia Rani; Yuhandri Yuhandri
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.v7i2.1300

Abstract

Traditional laboratory techniques for examining the nutrient content of organic fertilizers, specifically nitrogen (N), phosphorus (P), and potassium (K), are expensive, time-intensive, and pose environmental hazards. To address these issues, this paper presents a novel, non-destructive, image-based classification algorithm to identify fertilizer nutrient content. The proposed technique integrates color space conversion, unsupervised clustering, texture extraction, and an adapted New Identification Weighting (NIW) method. The NIW is derived from prior probability-based distance measurements and optimized with a balancing weighting factor to improve analytical stability across heterogeneous agricultural images. First, RGB images of fertilizers are converted into the perceptually uniform CIE L*a*b color space, which enhances color distinction under varying lighting conditions. Next, the images are segmented using K-Means clustering, followed by Gray-Level Co-occurrence Matrix (GLCM) extraction to capture textural and structural features. A key innovation of this research is the NIW method, functioning as an adaptive feature prioritization tool that assesses each features contribution to nutrient classification, effectively overcoming the limitations of previous a priori approaches. The system was tested on a dataset of 500 organic fertilizer images, achieving an overall classification accuracy of 97%, demonstrating its effectiveness and robustness. This approach offers a highly accurate and interpretable alternative to conventional chemical testing, making it a feasible, scalable, and affordable field tool for smart farming. By enabling on-site nutrient analysis, it strongly supports sustainable agricultural practices. Future work will focus on enhancing the systems flexibility to varying environmental conditions and integrating this approach into mobile-based diagnostic devices to facilitate real-time decision-making in agriculture.
Problems, Challenges, and Opportunities Visualization on Big Data Tri Wahyuningsih
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.v1i1.8

Abstract

Today, almost everything is recorded digitally, from our browsing history to our health records in hospitals, we make and process billions of data every day. In this era of big data, large amounts of data are continuously obtained for different purposes. But, just processing and analyzing the data isnt enough. If data is displayed visually, humans always search for patterns more effectively. Visualization and interpretation of the data are very critical tasks in making choices in various industries. This also guides us to new ways to find innovative ideas through visualization to solve big-data problems. In this paper, we will discuss the problems, challenges, and potential of Visualization in Big Data.
Classification of Batik Motifs Using Multi-Texton Co-Occurrence Descriptor and Binarized Statistical Image Features Ahmad Rizki Maulana; Suprapto Suprapto; Dyah Aruming Tyas
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.v6i2.622

Abstract

This study aims to enhance the classification accuracy of batik motifs through a novel integration of Multi-Texton Co-Occurrence Descriptor (MTCD) and Binarized Statistical Image Features (BSIF). The primary objective is to develop a robust feature extraction method that effectively captures both textural and statistical properties of batik images, specifically utilizing the Batik Nitik 960 dataset. Our methodology employs a combination of MTCD and BSIF, followed by Principal Component Analysis (PCA) for dimensionality reduction, optimizing the model's ability to learn from diverse characteristics inherent in batik motifs to augment the diversity and robustness of the training data, we enhanced the Batik Nitik 960 dataset by applying vertical flipping, in addition to existing rotations. We explored three feature fusion approaches: Combination 1, where features are combined before normalization and PCA, achieving an accuracy of 99.948%; Combination 2, where normalization occurs prior to feature combination, also achieving an accuracy of 99.948%; and Combination 3, which applies PCA separately to each feature before combination, resulting in an accuracy of 99.896%. Experimental results demonstrate a remarkable accuracy in classifying these motifs, with the combined MTCD-BSIF features significantly surpassing the individual performances of MTCD at 95.729% and BSIF at 99.531%. This substantial improvement addresses the limitations identified in previous research, which reported an accuracy of only 0.71 on the same dataset. Furthermore, we explore the impact of various feature fusion techniques on classification performance, providing insights into the effectiveness of our proposed methods. Our findings suggest that the combined MTCD-BSIF approach can serve as a benchmark for future studies aiming to enhance classification accuracy in similar domains, thereby contributing to advancements in automated classification systems and their applications across various fields.
Performance Improvement of Covid-19 Cough Detection Based on Deep Learning with Segmentation Methods Suyanto Suyanto; Zanjabila Zanjabila; Bagus Tris Atmaja; Wiratno Argo Asmoro
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.v5i2.205

Abstract

COVID-19 is an emergency problem that is being widely discussed in the world, one of which is the deep learning-based COVID-19 detection method which has been developed based on images of the patient's chest or cough. In this research, we propose a way to improve the performance of deep learning-based COVID-19 cough detection by using a segmentation method to produce several audio files containing one cough signal from one audio file containing several cough sound signals. In addition, we enabled two automatic cough segmentation methods, namely a Hysteresis Comparator based on the power spectrum and an RMS threshold based on the RMS energy value. The results obtained show that using the segmentation method for cough sounds can improve the model's performance in detecting COVID-19 coughs by 4% to 8%. The segmentation process can also remove noise between cough sound signals and provide a standard input model in the form of one cough signal. In addition, the segmentation results show information related to the characteristics of COVID-19 cough. The evaluation results show that the hysteresis comparator method has better results with an unweighted accuracy (UA) value of 83.19% compared to the RMS threshold method with a UA value of 79.06%.