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
The Integration of DEMATEL and SAW Methods for Developing a Research Performance Assessment Model for Lecturers Muh. Nurtanzis Sutoyo; Alders Paliling
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.v6i2.550

Abstract

This work aims to integrate two decision analytic methodologies, DEMATEL and SAW, to develop a comprehensive and effective model for assessing research performance among instructors. These strategies aim to rectify the deficiencies of traditional evaluation models, which often neglect the complexity of interconnections among performance metrics. This study utilizes research performance data from lecturers, encompassing publication count, journal quality, impact, funding, and cooperation. SAW is employed to calculate aggregate scores utilizing weights obtained from the DEMATEL analysis, whereas DEMATEL is utilized to delineate and assess the interrelationships among the evaluation criteria. The results indicate that the quantity of publications significantly influences research quality, succeeded by research impact and journal quality. Alternative A, with a maximum score of 0.996, demonstrated that the professor excelled in nearly all categories. A clear and objective evaluation methodology was developed by integrating DEMATEL with SAW. The development of more flexible criterion weights to accommodate shifts in academic practices and research priorities is a significant implication for future investigations. To evaluate this model's appropriateness and effectiveness in various academic contexts, it must be further assessed across multiple topic areas and types of educational institutions. This study facilitates the implementation of big data technology in academic performance evaluation, enhancing the accuracy and relevance of assessment methods.
Comparative Analysis of SVM and RF Algorithms for Tsunami Prediction: A Performance Evaluation Study Husni Teja Sukmana; Yusuf Durachman; Amri Amri; Supardi Supardi
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.v5i1.159

Abstract

This study explores the use of machine learning algorithms, specifically SVM and RF, for predicting tsunamis, a crucial aspect of disaster management. The research utilized earthquake data from 2001 to 2023, evaluating these models based on accuracy, precision, recall, F1-score, and ROC AUC, emphasizing features like magnitude, depth, and alert levels. The SVM model demonstrated an accuracy of 65.61%, precision of 70.59%, recall of 19.67%, F1-score of 30.77%, and ROC AUC of 62.15%. In comparison, the RF model showed an accuracy of 61.15%, precision of 50.00%, higher recall of 36.07%, F1-score of 41.90%, and ROC AUC of 63.84%. These results highlight the distinct strengths of each model: SVM's precision makes it suitable for minimizing false positives, while RF's higher recall indicates its effectiveness in detecting actual tsunamis. The findings underscore the significance of selecting the appropriate model for tsunami prediction based on specific disaster management needs and the inherent trade-offs in model selection. The research concludes that SVM and RF models provide valuable yet distinct contributions to tsunami prediction. Their application should be customized to disaster management requirements, balancing accuracy and operational efficiency. This study contributes to disaster risk management and opens avenues for further research in enhancing the accuracy and reliability of machine learning in natural disaster prediction and response systems.
AMIKOM-RECSYS: Enhancing Movie Recommender System using Large Language Model (ChatGpt), Deep Learning and Probabilistic Matrix Factorization Hanafi Hanafi; Anik Sri Widowati; Sri Ngudi Wahyuni
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.v6i4.897

Abstract

E-commerce has become one of the most widely used digital applications globally, enabling personalized product discovery and purchasing. To support these services, recommender systems are essential, offering item suggestions based on user preferences. Most recommender systems rely on machine learning algorithms to estimate user-item relevance scores, often utilizing product ratings. However, a persistent challenge in this domain is the issue of data sparsity, where only a small fraction of users provides explicit ratings, leading to reduced accuracy in recommendation results. In this study, we introduce a novel hybrid recommendation algorithm, named AMIKOM-RECSYS, designed to address the sparsity problem and enhance rating prediction. Our model integrates three main components included a Large Language Model (LLM) using ChatGPT, a Transformer-based encoder (BERT), and Probabilistic Matrix Factorization (PMF). The LLM generates descriptive information about movies based on specific prompts, which is then passed to BERT to encode the content into meaningful 2D vector representations. These enriched embeddings are subsequently utilized by the PMF algorithm to predict missing user-item ratings. We evaluate the proposed model on two benchmark datasets, ML-1M and ML-10M using Root Mean Squared Error (RMSE) as the evaluation metric. The AMIKOM-RECSYS model achieved RMSE values of 0.8681 on ML-1M and 0.7791 on ML-10M under a 50:50 data split, outperforming several baseline models including CNN-PMF, LSTM-PMF, and Attention-PMF. These results highlight the effectiveness of integrating LLM and Transformer-based contextual understanding into matrix factorization frameworks. In future work, we plan to extend this framework by incorporating other matrix factorization techniques such as Singular Value Decomposition (SVD) and integrating additional sources of user information, including social media activity, to further improve recommendation performance.
Developing a Parallel Network Slack-Based Measure Model in the Occurrence of Hybrid Integer-Valued Data and Uncontrollable Factors Syarifah Nurfuaduz Zakiah Habib Dzulkarnain; Mohd Kamal Mohd Nawawi; Rosmaini Kashim
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.v5i4.407

Abstract

This study develops an alternative approach to the parallel network Slack-Based Measure (SBM) Data Envelopment Analysis (DEA) model, offering a more accurate and informative assessment of performance within a network system. Traditional DEA models solely focus on the input utilization and the outputs produced when assessing efficiency, disregarding the operation of internal processes within a network system. In addition, these approaches do not assess the concurrent requirement of hybrid integer-valued data and uncontrollable factors on efficiency measures. To address these gaps, we propose a novel approach to parallel network SBM DEA model that integrates hybrid integer-valued data with uncontrollable factors, aiming for a more precise evaluation. Both requirements were initially integrating into the existing method. Subsequently, the optimal solution for the proposed method was achieved by converting its fractional form into a linear one. Therefore, the measures of the proposed approach can now deal directly with controllable hybrid integer- valued input and output slacks. We applied this model to a dataset of 26 faculties in a Malaysian public university, followed by a comparative analysis with existing models. Empirical findings indicate that four (4) faculties are found to be overall effective, as all of their internal processes are effective, while the other faculties are ineffective since not all of their internal processes are effective. The results from our model enable decision-makers to identify ineffectiveness within network processes, thereby facilitating targeted improvements in system performance. By concentrating on the appropriate processes, management can enhance their overall effectiveness and internal effectiveness.
Performance Evaluation of Support Vector Machine (SVM) and XGBoost for Predicting Toddlers’ Stunting Status Based on Anthropometric Data Nurjoko Nurjoko; Admi Syarif; Favorisen R. Lumbanraja; Khairunisa Berawi
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.v7i2.1260

Abstract

Stunting remains a primary global health concern, particularly in developing countries, due to its long-term effects on physical growth, cognitive development, and overall well-being. Despite various public health initiatives, challenges in early detection persist, highlighting the need for accurate, data-driven predictive models to support targeted interventions. This study aims to develop and compare the performance of two machine learning algorithms—SVM and Extreme Gradient Boosting (XGBoost)—for classifying stunting status among children under five, in order to determine the most effective method for early prediction. A quantitative machine learning approach was applied to a dataset comprising 17,498 records derived from Posyandu data in Lampung Province, Indonesia. The analytical pipeline included data preprocessing, class rebalancing using the Synthetic Minority Over-sampling Technique (SMOTE), and model evaluation through stratified 10-fold cross-validation. Performance was assessed using accuracy, precision, recall, and F1-score. The XGBoost model demonstrated superior performance with accuracy, precision, recall, and F1-score reaching 0.9979. In comparison, the SVM model produced slightly lower yet still strong results, achieving an accuracy of 0.9949, with similarly consistent performance across other evaluation metrics. These findings indicate that XGBoost more effectively handles high-dimensional, imbalanced data and captures nonlinear patterns in the dataset. XGBoost was identified as the optimal method for stunting classification in this study, outperforming SVM across all evaluation metrics. These results support the integration of boosting-based models into early detection systems for child nutritional assessment. Future studies should incorporate additional environmental and socioeconomic variables and evaluate model applicability in a real-time community health setting.
Predicting Dropout on E-learning Using Machine Learning Akmal Akmal
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.v1i1.9

Abstract

After the corona virus outbreak (COVID-19) many various institutions changed how they work. From various sectors, which experienced the biggest change was the education sector. The education sector, which in general requires face-to-face interaction between teachers and students in a place, has now changed to online, which does not require that both parties be in a place. This is certainly a very big change and has an impact. In this paper we will discuss e-learning methods for drop-out prediction, based on three techniques of machine learning.
Unveiling Hybrid Model with Naive Bayes, Deep Learning, Logistic Regression for Predicting Customer Churn and Boost Retention Devibala Subramanian; Ajitha Ajitha; Siti Sarah Maidin
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.v6i2.675

Abstract

The telecommunications sector is rapidly evolving but is increasingly challenged by customer churn, where subscribers switch to competing service providers. This study introduces a hybrid model for churn prediction and customer retention by combining machine learning methods—Naive Bayes, Deep Learning, and Logistic Regression—with sentiment analysis on user-generated content (UGC). Data was gathered through two primary sources: survey responses and 352 social media comments from users aged 20–35. The survey data was enriched with features such as gender, age, subscription period, complaints, and retention efforts. The preprocessing steps included handling missing values, scaling features, and encoding categorical variables to ensure model robustness. Experimental results demonstrated that Logistic Regression achieved the highest accuracy (88.45%) and sensitivity (91.33%) in detecting potential churners. The PCA-based approach followed closely with an accuracy of 86.77% and a balanced sensitivity-specificity profile (89.95% and 83.58%, respectively), effectively capturing key churn indicators. Random Forest and Decision Tree classifiers yielded lower sensitivity but remained strong in specificity, indicating their suitability for identifying loyal customers. Attribute weight analysis across models revealed that subscription plan, age, and retention effort were consistently influential in churn prediction. Furthermore, the integration of sentiment analysis provided emotional context to churn behavior, with negative comments triggering alerts for proactive engagement. The study highlights the predictive strength of combining structured survey data and unstructured UGC through machine learning and sentiment analytics. It underscores the importance of personalized retention strategies based on model interpretability and correlation weight findings. This hybrid approach equips telecom companies with actionable insights to minimize churn and sustain customer loyalty in a competitive market.
Teachable Machine: Optimization of Herbal Plant Image Classification Based on Epoch Value, Batch Size and Learning Rate Edwin Ariesto Umbu Malahina; Mardhalia Saitakela; Semlinda Juszandri Bulan; Marinus Ignasius Jawawuan Lamabelawa; Yohanes Suban Belutowe
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.v5i2.206

Abstract

Herbal plants are a source of natural materials used in alternative medicine and traditional therapies to maintain health. The purpose of this research is to develop an intelligent system application that is able to assist people in independently detecting herbal plants around them, provide education, and most importantly, find the optimal value based on certain parameters. This research uses several values for the parameters studied, namely the epoch value which varies between 10, 50, 100, 250, 750, and 1000; the batch size value which varies between 16, 32, 64, 128, 256, and 512; and the learning rate value which varies between 0.00001, 0.0001, 0.001, 0.01, 0.1, and 1. A total of 10,000 training data samples (1,000 samples in 10 classes) were used in Teachable Machine. The method used is to utilize the TensorFlow framework in the Teachable Machine service to train image data. This framework provides Convolutional Neural Networks (CNN) algorithms that can perform image classification with a high degree of accuracy. The test results for more than three months showed that the highest optimal value was achieved at the 50th epoch value, with a learning rate of 0.00001, and a batch size of 32, which resulted in an accuracy rate between 98% and 100%. Based on these results, a mobile web-based intelligent system application service was developed using the TensorFlow framework in Teachable Machine. This application is expected to be widely implemented for the benefit of the community. However, the challenges and limitations in training this test data are the large number of data classes that will be very good so that machine learning can learn to recognize objects but will take hours to train, then the training image object data has a clean background from other objects so that when tested it is not detected and influenced as another object or can result in a decrease in the percentage value.
Hybrid Transformer-XGBOOST Model Optimized with Ant Colony Algorithm for Early Heart Disease Detection: A Risk Factor-Driven and Interpretable Method Moch Deny Pratama; Faris Abdi El Hakim; Dimas Novian Aditia Syahputra; Dodik Arwin Dermawan; Asmunin Asmunin; Salamun Rohman Nudin; Andi Iwan Nurhidayat
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.v7i1.969

Abstract

Cardiovascular diseases (CVDs) remain the leading cause of death worldwide, with significant socioeconomic consequences due to premature death and chronic disability. Although clinical screening techniques have evolved, early and accurate prediction of heart disease is still partial due to the limited capacity of conventional machine learning algorithms to model the complex nonlinear interactions among various contributing risk factors e.g., hypertension, diabetes, hyperlipidemia, and genetic predisposition. To address these challenges, this research introduces a hybrid framework that combines the Transformer architecture known for its robust self-attention mechanism and high representational capabilities with Ant Colony Optimization (ACO), a nature-inspired metaheuristic algorithm modeled on the foraging behavior of ants, to enable adaptive and efficient hyperparameter optimization. The proposed model processes structured clinical data by encoding categorical variables into embeddings and normalizing numerical features, resulting in a unified tabular representation suitable for transformer-based analysis. ACO improves model efficiency by optimizing key parameters e.g., embedding configuration, learning rate, and depth, reducing manual intervention and computational overhead. The proposed Hybrid Transformer-ACO model focuses on interpretable clinical features to provide actionable risk stratification. Model evaluation was performed using classification metrics e.g., accuracy, precision, recall, F1 score, and time complexity to measure predictive performance and computational efficiency during the training and inference phases. These evaluation criteria provide evidence of the model's diagnostic reliability, generalizability, and practical feasibility for clinical application.. The model achieved 100% accuracy, sensitivity, specificity, and F1-score, outperforming several models. Time complexity analysis demonstrated efficient training and testing, while the model interpretability supports transparency and trust.
Classification of Starling Images Using a Bayesian Network April Lia Hananto; Aviv Yuniar Rahman; Tukino Paryono; Bayu Priyatna; Agustia Hananto; Baenil Huda
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.v6i1.423

Abstract

The classification of starling species is vital for biodiversity conservation, especially as some species are endangered. This research investigates the effectiveness of the Bayesian Network (BayesNet) for classifying starling species and compares its performance with Artificial Neural Networks (ANN) and Naive Bayes models. The dataset comprises 300 images of five starling species—Bali, Rio, Moon, Kebo, and Uret—captured under controlled conditions. Feature extraction focused on color, texture, and shape, while data augmentation through slight image rotations was applied to enhance model generalization. The BayesNet model achieved an accuracy of 96.29% using a 90:10 training-to-testing split, outperforming ANN (90.74%) and Naive Bayes variants. Precision, recall, F1-score, and AUC-ROC values further validated the robustness of the BayesNet model, with precision at 0.90, recall at 0.91, F1-score at 0.92, and AUC-ROC at 0.95. These results demonstrate the superior performance of multi-feature Bayesian Networks in starling classification compared to other machine learning models. The novelty of this study lies in its application of a probabilistic approach using Bayesian Networks, which enhances interpretability and performance, especially in scenarios with limited data. Future work may explore additional feature sets and advanced machine learning models to further improve classification accuracy and robustness.

Page 82 of 127 | Total Record : 1270