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
Revolutionizing Digit Image Recognition: Pushing the Limits with Simple CNN and Challenging Image Augmentation Techniques on MNIST Khodijah Hulliyah; Normi Sham Awang Abu Bakar; Saepul Aripiyanto; Dewi Khairani
Journal of Applied Data Sciences Vol 4, No 3: SEPTEMBER 2023
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jads.v4i3.104

Abstract

This study aims to apply Convolutional Neural Networks (CNN) and image augmentation techniques in digit recognition using the MNIST dataset. We built a CNN model and experimented with various image augmentation techniques to improve digit recognition accuracy. The results showed that the use of CNN with image augmentation techniques was effective in improving digit recognition performance. In the data collection stage, we used the MNIST dataset consisting of images of handwritten digits as training and testing data. After building the CNN model, we apply image augmentation techniques such as rotation, shift, and flipping to the training data to enrich the data variety and prevent overfitting. The evaluation results show that the CNN model that has been trained with image augmentation techniques produces significant accuracy, with a maximum accuracy of 99.81%. We also performed an ensemble of several CNN models and found that this approach increased the digit recognition accuracy to 99.79%. This research has the potential for further development. Recommendations for further research include exploring more specific and complex image augmentation techniques, as well as using more challenging datasets. In addition, future research may consider improvements to the CNN architecture used or combining it with other methods such as recurrent neural networks (RNN).
Lora Communication System for Early Detection and Monitoring of Water Toxicity in Floating Net Cages Rahmafadilla Rahmafadilla; Indrarini Dyah Irawati; Mochammad Fahru Rizal; Siti Sarah Maidin
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.v6i3.787

Abstract

Floating Net Cages/ Keramba Jaring Apung (KJA) are at risk of polluting the air, which can affect fish farming. Therefore, an early monitoring system is needed that can measure air quality such as temperature, pH, and dissolved oxygen (DO) in real-time. This system utilizes the LoRa RFM95W module to wirelessly transmit environmental data from sensors installed on the cages, which continuously monitor water quality parameters such as temperature, pH, and DO in real-time. The data obtained is then processed to monitor changes in water toxicity in real-time, allowing early detection of potential threats to the ecosystem. Tests were conducted at distances of 50m, 180m, 300m, 340m, and 440m. The results showed that the system worked well up to a distance of 300m with RSSI values between -85 dBm to -120 dBm and SNR more than 2 dB. However, at distances of 340m and 440m, the signal decreased and the delay increased. At a depth of 340m, only one experiment was successful with RSSI -134 dBm and SNR -6 dB, while at a depth of 440m, only a few experiments were successful with RSSI between -122 dBm to -132 dBm and SNR between 1 dB to -6 dB. The prototype system successfully transmitted real-time air quality data to a web-based monitoring center. Data from the sensors were sent via the LoRa network to a central server for further monitoring.
Ethno-Flipped Learning of Mathematics Lessons Based on Internalizing Data from Tri Mandala Concept on Schoology Platform I Made Ardana; I Made Sugiarta; I Gde Wawan Sudatha; Made Susi Lissia Andayani; Dewa Gede Hendra Divayana
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.v5i3.361

Abstract

In the digital era like today, maximum Mathematics learning outcomes in the affective and psychomotor domains are difficult to obtain through full online learning. This difficulty is influenced by environmental and cultural factors inherent in students’ lives. Therefore, an idea in the form of innovative learning model is needed to internalize cultural values and local wisdom in online learning. Based on the problems and needs of their solutions, the innovative model that emerged in this study is called Ethno-Flipped learning of Mathematics lessons based on internalizing data from Tri Mandala concept on the Schoology platform. The aim of this research is to demonstrate an Ethno-Flipped Learning design based on internalizing data from the Tri Mandala concept on the Schoology platform. This research approach is development with a focus on development at several stages, including: research and field data collection, planning, design development, initial trials, and revision of initial trial results. The number of personnel involved in the initial trial was 104 respondents. The data collection tool for the initial trial results was a questionnaire consisting of 10 questions related to the model design. The analysis was carried out by comparing the percentage of model design quality with the five scale categorization standards. The findings or research results show that the Ethno-Flipped learning design especially for Mathematics lessons based on internalizing data from the Tri Mandala concept on the Schoology platform is classified as good quality by a quality percentage of 87.54%. The contribution of this research is that it provides new knowledge about an innovative learning model based on Balinese culture and local wisdom combined with an online learning platform. The novelty of this research is in the form of an innovative learning model design that combines the Flipped Classroom learning model, the Ethno-Mathematics approach, and the Tri Mandala concept.
Comparison of Multilingual Model Sensitivity for Political Fact Verification with Integrated Multi-Evidence Nova Agustina; Kusrini Kusrini; Ema Utami; Tonny Hidayat
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.v7i2.1198

Abstract

Political news is frequently targeted by the dissemination of fake news on social media, which can influence public opinion and undermine trust in democratic processes. The main challenge in addressing this issue lies in the limited sensitivity of cross-lingual fact verification models in capturing semantic relationships between claims and evidence in long-text, multi-evidence settings. Existing approaches often struggle to assess the relevance and quality of evidence, resulting in suboptimal verification performance. This study compares three multilingual Large Language Models (LLMs), namely mBERT, XLM-R, and LaBSE, for political fact verification using an integrated multi-evidence approach. Experiments are conducted on the PolitiFact dataset, with performance evaluated using sensitivity, accuracy, precision, and F1-score metrics.The results indicate that mBERT achieves the highest overall sensitivity at 89.44%, followed by LaBSE at 81.81% and XLM-R at 78.81%. However, mBERT exhibits lower precision, whereas LaBSE provides a better balance between precision (87.02%) and accuracy (86.46%), resulting in an F1-score of 84.33%. XLM-R demonstrates lower sensitivity but maintains competitive precision (85.47%) and accuracy (84.60%), with an F1-score of 82.00%. Sensitivity analysis based on the number of evidence reveals distinct model behaviors, where mBERT performs optimally with six pieces of evidence, XLM-R is more effective under limited evidence conditions, and LaBSE shows a stable and increasing sensitivity trend as the amount of evidence increases, indicating robustness in multi-evidence scenarios. Further statistical analysis shows that XLM-R has the lowest performance variance, while LaBSE statistically outperforms mBERT in several evaluation aspects. Overall, LaBSE is recommended as the most balanced model for multi-evidence-based political fact verification.
Improving Early Detection of Cervical Cancer Through Deep Learning-Based Pap Smear Image Classification Nita Merlina; Arfhan Prasetio; Ida Zuniarti; Nissa Almira Mayangky; Daning Nur Sulistyowati; Faruq Aziz
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.v6i2.576

Abstract

Cervical cancer is one of the leading causes of death in women worldwide, making early detection of the disease crucial. This study proposes a deep learning-based approach that has the advantage of leveraging pre-trained models to save data, time, and computation to classify Pap smear images without relying on segmentation, which is traditionally required to isolate key morphological features. Instead, this method leverages deep learning to identify patterns directly from raw images, reducing preprocessing complexity while maintaining high accuracy. The dataset used in this study is a public data repository from Nusa Mandiri University (RepomedUNM), which has a wider variety of data. This dataset is used to classify images into four categories: Normal, LSIL, HSIL, and Koilocytes. The dataset consists of 400 images evenly distributed, ensuring class balance during training. Transfer learning is applied using five Convolutional Neural Network (CNN) architectures: ResNet152V2, InceptionV3, ResNet50V2, DenseNet201, and ConvNeXtBase. To prevent overfitting, techniques such as data augmentation, dropout regularization, and class weight adjustment are applied. The evaluation results in this study showed the highest accuracy with a value of ResNet152V2 = 0.9025, InceptionV3 = 0.8953 and DenseNet201 = 0.8845. ResNet152V2 excelled in extracting complex features, while InceptionV3 showed better computational efficiency. The study also highlighted the clinical impact of misclassification between Koilocytes and LSIL, which may affect diagnostic outcomes. Data augmentation techniques, including horizontal and vertical flipping and normalization, improved the model's generalization to a wide variety of images. Specificity was emphasized as a key evaluation metric to minimize false positives, which is important in medical diagnostics. The findings confirmed that transfer learning effectively overcomes the limitations of small datasets and improves the classification accuracy of pap smear images. This approach shows potential for integration into clinical workflows to enable automated and efficient cervical cancer detection.
Information Security Measurement using INDEX KAMI at Metro City Ratna Savitri; Firmansyah Firmansyah; Dworo Dworo; Muhammad Said Hasibuan
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.v5i1.152

Abstract

Information security is a crucial issue that affects the overall business process, therefore it must be protected and secured. This research was conducted to assess the information security risks at Metro City Communication and Information Office in a structured manner towards information assets in identifying efforts to reduce risks as part of the information security management program. The research method begins with defining the scope, collecting data and supporting documents, evaluating the Information Security Index (KAMI), determining scores in 7 security areas, where strengths/maturity and weaknesses/deficiencies will be identified in each security area. Finally, after obtaining the evaluation results, recommendations will be made. The Information Security Index (KAMI) is a computer-based tool in excel format that can assess and evaluate the completeness and maturity level of information security implementation based on the SNI ISO/IEC 27001 criteria that describe the readiness of the information security framework. The data obtained by the researcher is based on interview results, examination of the availability of Information Security Management System (SMKI) documents, and evidence of SMKI implementation records/archives. The dashboard evaluation results for electronic system category score 17, which is in the high category, governance score is 69, risk management score is 29, framework score is 33, information asset management score is 69, technology score is 81 and supplement score is 0%. Based on verification of the results of the KAMI Index version 4.2 assessment file, a score of 275 was obtained, indicating that information security
Clustering Cadet Training Performance Using K-Means and Ward's Method Evidence from OTMon Maritime Monitoring System Dian Wahdiana; Ekohariadi Ekohariadi; Ratna Suhartini
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.v6i4.958

Abstract

This study investigates cadet performance segmentation during on-board maritime training using clustering analysis of data from the On Training Monitoring (OTMon) system. Grounded in the competency-based education framework and experiential learning theory, the research aims to identify behavioral patterns and competency levels among 80 maritime cadets over a twelve-month sea-based training program. The OTMon application continuously recorded task completion rates, feedback interactions, sign-on consistency, and report submissions. K-Means clustering and Principal Component Analysis (PCA) revealed three distinct cadet profiles: Cluster 1 (high-performing) with average task completion of 92.4% and feedback frequency of 15.2 times/month; Cluster 2 (administratively consistent) with 88.1% completion but only 6.3 feedback interactions/month; and Cluster 3 (at-risk) with 67.5% completion and 3.8 feedback interactions/month. Linear Discriminant Analysis (LDA) validated the clusters with 98.8% resubstitution accuracy and 97.6% cross-validation accuracy, supported by generalized squared distances above 9.5 between all cluster pairs, indicating strong separation. These findings demonstrate that unsupervised clustering can reliably distinguish high-performing cadets from those needing targeted intervention, enabling data-informed mentoring and adaptive learning strategies in maritime education. The contribution of this study lies in integrating digital monitoring data with both unsupervised and supervised machine learning methods to enhance competency assessment. The novelty is in applying maritime-specific learning analytics for real-time performance segmentation, offering a scalable diagnostic framework for improving supervision quality and supporting individualized cadet development in vocational training contexts.
The Efficacy of Online Gamification in Improving Basic English Skills for Fourth-Grade Students Tiamyod Pasawano; Thosporn Sangsawang
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.v5i4.410

Abstract

The study aimed to achieve three main objectives: 1) to develop an online gamification system using digital learning platforms for teaching English to Grade 4 students, following the E1/E2 = 80/80 efficiency criterion, 2) to compare students' achievement in Basic English through online gamification, and 3) to assess students' satisfaction with the use of online gamification in learning Basic English. The sample comprised 30 Grade 4 students from Settabutr Upathum School in the academic year 2022, selected through purposive random sampling. Research instruments included online Zoom classes, lesson plans, and interactive learning platforms. The study employed mean, standard deviation, and t-tests for dependent samples for data analysis. The results revealed an efficiency value of E1/E2 as 70.00/69.00, falling short of the 80/80 criteria. Several factors, such as the comprehensive nature of testing macro skills using digital media beyond cognitive abilities, may have contributed to not meeting the set criterion. Furthermore, a significant improvement in learning achievements in Basic English was observed among Grade 4 students who used online gamification compared to traditional methods, with higher scores in achievement tests at a significance level of 0.05. Finally, students expressed a good level of satisfaction with the online gamification approach in learning Basic English.
Optimized Hybrid CNN-LSTM Model for Predicting Transportation Sector Stock Prices Using Optimizer and Activation Function Tuning Fiena Gunawan; Dinar Ajeng Kristiyanti
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.v7i2.1323

Abstract

Stock price prediction is inherently complex due to nonlinear dynamics and high volatility, particularly in Indonesia’s transportation sector, which experienced significant inflationary pressure and extreme instability during and after the COVID-19 pandemic. These disruptions introduced structural breaks and regime shifts, intensifying non-stationary market behavior and increasing forecasting uncertainty. Such conditions create an urgent need for robust predictive information systems capable of supporting investment decision-making and risk management in highly volatile environments. However, standalone recurrent models such as Long Short-Term Memory (LSTM) often struggle to capture local micro-patterns and long-term dependencies. Moreover, prior studies have rarely implemented systematic hyperparameter optimization, resulting in inconsistent predictive performance across stocks with heterogeneous volatility. In contrast, Convolutional Neural Networks (CNN) extract local patterns and short-term nonlinear features, making them effective for modeling high-frequency fluctuations. This study proposes a systematically optimized hybrid CNN-LSTM model to forecast transportation sector stock prices using daily OHLC data from 2020-2025. The research framework follows the Cross Industry Standard Process for Data Mining (CRISP-DM) methodology, encompassing business understanding, data understanding, data preparation, modeling, evaluation, and deployment. Prior to modeling, preprocessing includes data cleaning, Min-Max normalization, and sliding window transformation to construct supervised learning sequences. CNN is employed to extract localized nonlinear features and reduce noise, while LSTM models long-term temporal dependencies. Model performance is evaluated using MAE, MSE, RMSE, MAPE, and R². Results show that the optimized CNN-LSTM model outperforms the baseline across all stocks. The highest R² of 0.9725 is obtained from one stock, indicating strong performance. In addition, the average R² improves from 0.8736 to 0.9483, an increase of 0.0747 (8.55%). The best results are achieved using ReLU with Adam and Nadam optimizers, demonstrating improved convergence and generalization. These findings highlight the effectiveness of optimized hybrid deep learning models for forecasting in nonlinear and non-stationary financial markets.
Harnessing the Power of Prophet Algorithm for Advanced Predictive Modeling of Grab Holdings Stock Prices Hery Hery; Calandra A. Haryani; Andree E. Widjaja; Aditya Rama Mitra
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.v5i2.181

Abstract

This study investigates the effectiveness of the Prophet algorithm in predicting Grab Holdings' stock prices dataset from Kaggle. By meticulously analyzing historical closing, high, low, and volume data, the research aims to uncover market patterns and gain insights into investor sentiment based on short-term forecasting. The findings reveal a dynamic trajectory for Grab Holdings' stock, characterized by significant fluctuations and evolving investor confidence. The stock reached a peak of $14 in early 2022, indicating optimism, but subsequently experienced a decline to $4 by late 2023, reflecting a shift in sentiment. Notably, 2023 witnessed heightened volatility compared to 2022, evident in more significant price swings and increased trading volume. The Prophet algorithm demonstrated promising potential for prediction better than traditional methods, which overlook the presence of seasonality or fail to adapt to evolving market conditions, leading to less accurate forecasts. The excellent performance of Prophet is indicated by a Mean Absolute Percentage Error (MAPE) of 10.45511%, a Mean Absolute Error (MAE) of 3.112026, and a Root Mean Squared Error (RMSE) of 3.516969. Compared to the traditional ARIMA, MAE and RMSE resulting from Prophet are much lower than their counterparts, which are 14.49675 and 16.079898, respectively. These widely used metrics suggest moderate accuracy in predicting future stock prices. This research offers valuable insights for investors that they can use to understand the trend of Grab Holdings' stock price and make more informed investment decisions regarding buying or selling opportunities. However, it is crucial to acknowledge the inherent limitations of such models and interpret results cautiously, considering the ever-changing dynamics of the financial market.

Page 70 of 127 | Total Record : 1270