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
Assessing Large Language Models for Zero-Shot Dynamic Question Generation and Automated Leadership Competency Assessment I Gusti Bagus Yogiswara Gheartha; Adiwijaya Adiwijaya; Ade Romadhony; Yusfi Ardiansyah
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.v7i1.970

Abstract

Automated interview systems powered by artificial intelligence often rely on fine-tuned models and annotated datasets, limiting their adaptability to new leadership competency frameworks. Large language models have shown potential for generating questions and assessing answers, yet their zero-shot performance, operating without task-specific retraining remains underexplored in leadership assessment. This study examines the zero-shot capability of two models, Qwen 32B and GPT-4o-mini, within a multi-turn self-interview framework. Both models dynamically generated questions, interpreted responses, and assigned scores across ten leadership competencies. Professionals representing the role of Digital Marketing and Account Manager participated, each completing two AI-led interview sessions. Model outputs were evaluated by certified experts using a structured rubric across three dimensions: quality of behavioral insights, relevance of follow-up questions, and fit of assigned scores. Results indicate that Qwen 32B generated richer insights than GPT-4o-mini (mean = 2.86 vs. 2.62; p less than 0.01) and provided more differentiated assessments across competencies. GPT-4o-mini produced more consistent follow-up questions but lacked depth in interpretation, often yielding generic outputs. Both models struggled with accurate scoring of candidate responses, reflected in low answer score ratings (Qwen mean = 2.35; GPT mean = 2.21). These findings suggest a trade-off between insight richness and scoring stability, with both models demonstrating limited ability to fully capture nuanced leadership behaviors. This study offers one of the first empirical benchmarks of zero-shot model performance in leadership interviews. It underscores both the promise and current limitations of deploying such systems for scalable assessment. Future research should explore competency-specific prompt strategies, fairness evaluation across demographic groups, and domain-adapted fine-tuning to improve accuracy, reliability, and ethical alignment in high-stakes recruitment contexts.
Implementation of Stacking Technique Combining Machine Learning and Deep Learning Algorithms Using SMOTE to Improve Stock Market Prediction Accuracy Ibnu Rasyid Munthe; Bhakti Helvi Rambe; Fauziah Hanum; Ade Trya Amanda; Anita Sri Rejeki Hutagaol; Richi andrianto
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.v5i4.421

Abstract

This study introduces a stacking technique that integrates machine learning (ML) and deep learning (DL) algorithms to enhance the accuracy of stock market trend predictions. The stacking model utilizes XGBoost and Random Forest as base models from the ML domain, while Logistic Regression and LSTM (Long Short-Term Memory) function as meta models to optimize predictive accuracy. A significant challenge in stock market data is class imbalance, where certain trends, such as stock price drops, are underrepresented. To mitigate this, we applied the Synthetic Minority Over-sampling Technique (SMOTE) to generate synthetic data for the minority class. This approach helps the model better capture patterns from the underrepresented data while preserving essential information from the majority class. The implementation of SMOTE, coupled with the stacking technique, yielded a substantial improvement in prediction accuracy. The results showed that the Random Forest algorithm achieved an accuracy of 85% with precision, recall, and F1-score all at 85%, while XGBoost and Logistic Regression achieved accuracies of 82% and 81% respectively. For the deep learning models, LSTM reached an accuracy of 83%, while the Stacking Meta Model with LSTM achieved an accuracy of 83% with slightly better precision and recall at 84%. The stacking model, with Logistic Regression as the meta model, ultimately achieved the highest accuracy of 86%, outperforming individual models such as SVM (Support Vector Machine), LSTM, Random Forest, and Logistic Regression (LR). These findings demonstrate the efficacy of combining SMOTE with stacking to address data imbalance and improve stock market predictions. The novelty of this study lies in the integration of advanced ML and DL models within a stacking framework to handle class imbalance in financial datasets. Future research will explore the deployment of this model in a real-time web-based application to support investor decision-making in stock market trend analysis.
GoogLeNetMP: A Development of GoogLeNet Architecture for Multi-Class Microplastic Classification in Subsurface Water Image Halifia Hendri; Yuhandri Yuhandri; Agung Ramadhanu
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.v7i3.1372

Abstract

Microplastic pollution has become a major environmental concern due to its persistence in marine ecosystems and its potential impact on aquatic organisms and human health. Automatic detection of microplastic particles in underwater environments remains challenging because of turbidity, low contrast, light distortion, and the visual similarity between microplastics and natural marine objects. This study proposes GoogLeNetMP, an enhanced GoogLeNet-based deep learning architecture for multi-class classification of subsurface marine images into four categories: primary microplastics, secondary microplastics, non-microplastics, and marine biota. The proposed framework integrates basic image preprocessing (resizing and noise reduction) with a modified GoogLeNetMP architecture designed to intrinsically handle fine-grained feature extraction under degraded conditions, thereby minimizing the reliance on complex external enhancement pipelines. A dataset of underwater images acquired from the coastal waters of Padang, Indonesia, was used for model development and evaluation. Experimental results show that GoogLeNetMP outperformed the standard GoogLeNet model, achieving 95.75% accuracy, 92.80% sensitivity, 97.00% specificity, and an F1-score of 92.06%. The proposed model also demonstrated more stable training convergence and better discrimination of visually challenging classes. The architecture is designed to internalize the robust feature extraction process, thereby minimizing the reliance on extensive external enhancement pipelines while maintaining standard normalization steps for input consistency. These findings indicate that GoogLeNetMP is a promising approach for AI-based marine pollution monitoring and decision support in sustainable coastal management.
Knowledge Management Strategy by Means of Virtualization in Covid-19 Eissa Mohammed Ali Qhal
Journal of Applied Data Sciences Vol 1, No 2: DECEMBER 2020
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v1i2.46

Abstract

In an era where companies, governmental programs, and the economy are knowledge-driven, they must understand the best ways to apply knowledge management to yield better results. Fortunately, technology is continuously evolving to better cope with knowledge management's arising challenges by innovating better problem-solving models. Cloud computing is one of the many technologies today that has revolutionized how different companies and economic sectors treat knowledge management by enabling this Knowledge's virtualization. The paper focuses on knowledge management transition to virtualization technology in supporting businesses during the COVID-19. Pandemic times despite many social restrictions being put in place to contain the virus. The text further presents the results of a study involving major firms in different sectors and how their applied virtualization technology is yielding results even during the toughest of times in any company or economy.
Applying Structural Equation Modelling to Examine the Impact of Environmental Management Accounting on Financial Nga Thi Hang Nguyen
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.v6i3.728

Abstract

Environmental management accounting has garnered significant attention from various stakeholders. Enterprises that effectively implement environmental management accounting not only contribute to sustainable development but also enhance organizational performance. This study aims to examine the relationship between environmental management accounting, green innovation, and the financial performance of small and medium-sized enterprises in the context of a developing country like Vietnam. A quantitative research approach was employed to analyze data collected through a structured survey. The dataset comprises responses from 151 small and medium-sized enterprises, with financial managers and management accountants serving as key informants. Data analysis was conducted using the Smart Partial Least Squares software. The findings reveal that environmental management accounting has a direct positive impact on financial performance and an indirect impact through the mediation of green process innovation. While green product innovation exerts a direct impact on financial performance, environmental management accounting appears to have no significant influence on green product innovation. Consequently, green product innovation does not function as a mediating variable in the relationship between environmental management accounting and financial performance. The results underscore the greater significance of green process innovation over green product innovation in driving improvements in the financial performance of small and medium-sized enterprises. These results significantly contribute to the relatively unexplored theoretical relationship between environmental management accounting, green innovation, and the financial performance of small and medium-sized enterprises. Furthermore, the study provides a practical foundation for managers to boost their organizations’ financial performance by practicing environmental accounting and integrating green innovation into business operations.
The Determinant Factors For The Issuance Of Central Bank Digital Currency (CBDC) In Malaysia Using Machine Learning Framework Normi Sham Awang Abu Bakar; Norzariyah Yahya; Norbik Bashah Idris; Engku Rabiah Adawiah Engku Ali; Jasni Mohamad Zain; Erni Eliana Khairuddin; Ahmad Firdaus Zainal Abidin; Sheikh Mohammad Tahsin Murtaj; Siti Sarah Maidin
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.v5i2.176

Abstract

In order to identify the factors influencing the establishment of the Centre Bank Digital Currency (CBDC) in Malaysia, this study leverages the machine-learning technique to determine the most critical factors leading to CBDC issuance in Malaysia. The overall Central Bank Digital Currency Project Index (CBDCPI) was selected as a target variable,while two machine learning algorithms, Random Forest and XGBoost were used to identify the determining variables. The accuracy obtained through the Random Forest is 83% and subsequently, 80% in XGBoost. This study explored a new research frontier by creating two machine-learning models that treated retail and wholesale CBDCPI as target variables. The data used in the process are gathered from various official sources such as the Bank for International Settlements (BIS), the International Monetary Fund (IMF), and the World Bank. The Circulation of Cash, Prevalence of Cryptocurrencies, Effect of CBDC on International Trade, the Search Interest, Financial Development Index, Innovation Value, and Trade Openness are some of the most critical factors determining whether CBDC will be issued in Malaysia. Generally, are identified as important factors determining whether CBDC will be issued in Malaysia. Eventually, the factors identified will be used to develop a framework for the implementation of CBDC in Malaysia.
CNN-LSTM with Multi-Acoustic Features for Automatic Tajweed Mad Rule Classification Nenny Anggraini; Yusuf Rahman; Achmad Nizar Hidayanto; Husni Teja Sukmana
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.v7i1.1062

Abstract

The rules of mad recitation in the Qur’an are a crucial aspect of tajwīd, governing the lengthening of vowel sounds that affect both meaning and recitational accuracy. Despite its importance, there is currently no reliable automatic system capable of classifying mad rules based on voice input. This study proposes a deep learning-based approach using a hybrid Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model to automatically classify mad rules from Qur’anic recitations. The research follows the CRISP-DM methodology, covering data understanding, preparation, modeling, and evaluation stages. Acoustic features were extracted from 3,816 annotated audio segments of Surah Al-Fātiḥah, combining Mel-Frequency Cepstral Coefficients (MFCC), Chroma, Spectral Contrast, and Root Mean Square (RMS) to represent phonetic and prosodic attributes. The CNN layers captured spatial characteristics of the spectrum, while LSTM layers modeled temporal dependencies of the audio. Experimental results show that the combination of all four features achieved an accuracy of 97.21%, precision of 95.28%, recall of 95.22%, and F1-score of 95.25%. These findings indicate that multi-feature integration enhances model robustness and interpretability. The proposed CNN-LSTM framework demonstrates potential for practical deployment in voice-based tajwīd learning tools and contributes to the broader field of Qur’anic speech recognition by offering a systematic, ethically grounded, and data-driven approach to mad classification.
Enhancing Online Batik Shopping Experience through Live Streaming Commerce and the LYFY Application Trianggoro Wiradinata; Wilbert Bryan Wibowo; Yustus Eko Oktian; Indra Maryati; Yosua Setyawan Soekamto
Journal of Applied Data Sciences Vol 6, No 1: JANUARY 2025
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v6i1.504

Abstract

Online batik shopping often results in buyer dissatisfaction due to discrepancies between product descriptions and the actual items received. Static images and text on e-marketplace platforms are insufficient to convey the intricate details of batik designs, leading to mismatches in customer expectations. To mitigate this issue, Live Streaming Commerce (LSC) features, such as those on Shopee Live, allow sellers to showcase products in real-time, providing more accurate representations. However, sellers face challenges in managing overwhelming volume of comments during live streams, making it difficult to prioritize important queries. LYFY, a comment management app developed to streamline these interactions, aims to address this problem by improving the quality of interaction between live streamers and prospective buyers through filtering important comments. This study examines the determinants affecting the adoption of LYFY by online batik vendors. The research integrates the Task-Technology Fit (TTF), Technology Acceptance Model (TAM), and Expectation-Confirmation Model (ECM) frameworks to evaluate LYFY's performance in fulfilling user requirements. Data were collected from 243 respondents with LSC experience, and the research model underwent evaluation through Partial Least Squares Structural Equation Modeling (PLS-SEM). The measurement model exhibited high reliability and validity, with values surpassing the suggested thresholds, thereby providing solid support for subsequent analysis. Key factors such as TTF, confirmation, perceived usefulness, ease of use, and satisfaction were examined to determine their impact on user adoption. The analysis revealed that TTF has the strongest influence on confirmation, perceived usefulness, satisfaction, and individual performance. Additionally, perceived ease of use and confirmation substantially influence continuance intentions and satisfaction. These results suggest that enhancing LYFY's task-technology fit and simplifying its user interface are crucial for improving user satisfaction and adoption. By addressing these areas, LYFY can better support live stream sellers, reduce product expectation discrepancies, and improve overall customer experience, particularly in the online batik market.
Fuzzy Expert System for Melamine Moulding Compound Dye Feasibility Identification: A Case study on Plastic Industry Jansen Wiratama; Santo Fernandi Wijaya; Samuel Ady Sanjaya; Florentina Kurniasari; Hendro Budiyanto; Ala Al Kafri; Nuttaphat Sukchitt
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.v7i3.1498

Abstract

This study develops and evaluates an explainable zero order Fuzzy Sugeno expert system for assessing the feasibility of Melamine Moulding Compound (MMC) dyes in the melamine industry. The objective is to support consistent dye assessment by converting expert knowledge of boiling point, processing time, and injection pressure into a structured decision model. Each input variable is represented by low, medium, and high fuzzy sets, while four domain experts contributed to the definition and review of the membership ranges, rule base, and feasibility outcomes. The model contains 27 rules with singleton consequents of 10, 20, and 30 for not feasible, conditionally feasible, and feasible decisions, respectively, with midpoint decision boundaries of 15 and 25. The model was evaluated using 23 historical process profiles and classified 10 samples as not feasible (43.48%), 8 as conditionally feasible (34.78%), and 5 as feasible (21.74%). MMC103, with a boiling point of 165 °C, processing time of 65 seconds, and pressure of 55 bar, activated eight rules and produced a Sugeno score of 26.67, resulting in a feasible classification. A ±5% sensitivity analysis with 10,000 random perturbations produced MMC103 scores from 25.86 to 27.29 and retained its feasible classification in all runs, while overall class stability across the 23 profiles reached 98.86%. Sugeno classifications agreed with Mamdani inference for 20 of 23 profiles and with a crisp rule baseline for 18 profiles. By combining expert knowledge formalization, sensitivity evaluation, comparative analysis, and web implementation, this study provides an explainable and traceable decision support approach for MMC dye feasibility assessment in melamine production.
Application of Adaptive UKF Algorithm in Multi-target Tracking and Positioning System Guofang Liu; Xiong Wang
Journal of Applied Data Sciences Vol 4, No 2: MAY 2023
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v4i2.81

Abstract

Adaptive filtering algorithm (FIR) is a design method of adaptive variable target tracking system based on probability density distribution model. The algorithm realizes the target movement in the global range by estimating the parameters of different regions in the image, which improves the real-time performance and effectiveness.