Article Per Year (5 Year)
p-Index From 2021 - 2026
0.947
P-Index
This Author published in this journals
All Journal Journal of Applied Data Sciences Journal of Digital Market and Digital Currency
Deshinta Arrova Dewi
Faculty of Data Science and Information Technology, INTI International University, Nilai, Malaysia
Computer Science & IT Control & Systems Engineering Decision Sciences, Operations Research & Management Economics, Econometrics & Finance

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
Deep Learning-Based Loan Approval Prediction Using Artificial Neural Network (ANN) and Feature Importance Analysis Sheeba Armoogum; Deshinta Arrova Dewi; Vinaye Armoogum; Nicolas Melanie; Tri Basuki Kurniawan
Journal of Digital Market and Digital Currency Vol. 3 No. 1 (2026): Regular Issue March 2026
Publisher : Bright Publisher

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.47738/jdmdc.v3i1.55

Abstract

The increasing demand for efficient and objective credit evaluation has motivated the adoption of artificial intelligence in financial decision-making. This study proposes a deep learning-based loan approval prediction model using an Artificial Neural Network (ANN) combined with feature importance analysis to enhance interpretability. The dataset, consisting of 2,000 loan application records with both financial and demographic attributes, was preprocessed through normalization and one-hot encoding to ensure consistent feature representation. The ANN model was trained using three hidden layers (64–32–16 neurons) with the ReLU activation function and optimized using Adam with early stopping to prevent overfitting. Experimental results demonstrate that the proposed ANN model achieves an accuracy of 92%, with a precision of 0.91, a recall of 0.93, and a ROC-AUC of 0.95, indicating excellent classification capability. The Permutation Feature Importance analysis revealed that Credit Score, Income, and Loan Amount are the most significant predictors influencing loan approval decisions. These findings confirm that the ANN model can capture complex non-linear relationships among financial attributes while maintaining transparency through explainable AI techniques. The proposed approach contributes both theoretically and practically by combining predictive power with interpretability, offering a reliable and explainable framework for automating loan evaluation in modern financial institutions.
Clustering the Unlabeled Data Using a Modified Cat Swarm Optimization Deshinta Arrova Dewi; Tri Basuki Kurniawan; Mohd Zaki Zakaria; Sheeba Armoogum
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.v5i3.349

Abstract

This paper presents a modified version of the Cat Swarm Optimization (CSO) algorithm aimed at addressing the limitations of traditional clustering methods in handling complex, high-dimensional datasets. The primary objective of this research is to improve clustering accuracy and stability by eliminating the mixture ratio (MR), setting the counts of dimensions to change (CDC) to 100%, and incorporating a new search equation in the tracing mode of the CSO algorithm. To evaluate the performance of the modified algorithm, five classic datasets from the UCI Machine Learning Repository—namely Iris, Cancer, Glass, Wine, and Contraceptive Method Choice (CMC)—were used. The proposed algorithm was compared against K-Means and the original CSO. Performance metrics such as intra-cluster distance, standard deviation, and F- measure were used to assess the quality of clustering. The results demonstrated that the modified CSO consistently outperformed the competing algorithms. For example, on the Iris dataset, the modified CSO achieved a best intra-cluster distance of 96.78 and an F-measure of 0.786, compared to 97.12 and 0.781 for K-Means. Similarly, for the Wine dataset, the modified CSO reached a best intra-cluster distance of 16399, surpassing K-Means which recorded 16768. In conclusion, the modifications introduced to the CSO algorithm significantly enhance its clustering performance across diverse datasets, producing tighter and more accurate clusters with improved stability. These findings suggest that the modified CSO is a robust and effective tool for data clustering tasks, particularly in high-dimensional spaces. Future work will focus on dynamic parameter tuning and testing the scalability of the algorithm on larger and more complex datasets.
Efficient Web Mining on MyAnimeList: A Concurrency-Driven Approach Using the Go Programming Language Muhammad Daffa Arviano Putra; Deshinta Arrova Dewi; Wahyuningdiah Trisari Harsanti Putri; Harry Tursulistyono Yani Achsan
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.v5i3.352

Abstract

Anime is a globally popular form of entertainment, with the industry experiencing rapid growth in recent years. Despite the wealth of anime data available on MyAnimeList, the largest community-driven platform for anime enthusiasts, existing publicly available datasets are often outdated and incomplete. This presents a challenge for data science research, as the increasing volume of anime information requires more efficient data extraction methods. This research aims to address this challenge by developing a concurrent web mining program using the Go programming language. Leveraging Go's concurrency capabilities, our program efficiently extracted anime data from MyAnimeList, iterating through anime pages from ID 1 to 52,991. To overcome potential issues like rate limits and server timeouts, we implemented a two-phase execution strategy. As a result, the program successfully gathered 23,105 anime records within 8.5 hours. The extracted data has been transformed into a comprehensive dataset and made publicly available in CSV format. This research demonstrates the effectiveness of concurrent web mining for large-scale data extraction and offers a valuable resource for future data-driven research in the anime industry.
Unveiling Criminal Activity: a Social Media Mining Approach to Crime Prediction Sheeba Armoogum; Deshinta Arrova Dewi; Vinaye Armoogum; Nicolas Melanie; Tri Basuki Kurniawan
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.v5i3.350

Abstract

Social media platforms have become breeding grounds for abusive comments, necessitating the use of machine learning to detect harmful content. This study aims to predict abusive comments within a Mauritian context, focusing specifically on comments written in Mauritian Kreol, a language with limited natural language processing tools. The objective was to build and evaluate four machine learning models—Decision Tree, Random Forest, Naïve Bayes, and Support Vector Machine (SVM)—to accurately classify comments as abusive or non-abusive. The models were trained and tested using k-fold cross-validation, and the Decision Tree model outperformed others with 100% precision and recall, while Random Forest followed with 99% accuracy. Naïve Bayes and SVM, although achieving 100% precision, had lower recall rates of 35% and 16%, respectively, due to imbalanced data in the training set. Pre-processing steps, including stop-word removal and a custom Kreol spell checker, were key in enhancing model performance. The study provides a novel contribution by applying machine learning in a Mauritian context, demonstrating the potential of AI in detecting abusive language in underrepresented languages. Despite limitations such as the absence of a Kreol lemmatization tool and incomplete coverage of Kreol spelling variations, the models show promise for wider application in social media crime detection. Future research could explore expanding this approach to other languages and domains of social media crimes.
Integrating Convolutional Neural Networks into Mobile Health: A Study on Lung Disease Detection Muhammad Said Hasibuan; R Rizal Isnanto; Deshinta Arrova Dewi; Joko Triloka; RZ Abdul Aziz; Tri Basuki Kurniawan; Ary Maizary; Anggawidia Wibaselppa
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.v6i3.660

Abstract

This study presents the development and evaluation of a Convolutional Neural Network (CNN) model for lung disease detection from chest X-ray images, complemented by a mobile application for real-time diagnosis. The CNN model was trained on a diverse dataset comprising images labeled as "NORMAL" and "PNEUMONIA," achieving an overall accuracy of 96%. Compared to traditional machine learning methods such as Support Vector Machine (SVM) and Random Forest, which typically achieve accuracies ranging from 85% to 92%, the proposed CNN model demonstrates superior performance in classifying lung conditions. The model achieved high precision (0.98) and recall (0.96) for pneumonia detection, as well as precision (0.89) and recall (0.95) for normal cases, ensuring both sensitivity and specificity in diagnostic performance. These results indicate that the model minimizes false positives and false negatives, which is crucial for reducing misdiagnoses and improving patient outcomes in clinical settings. To enhance accessibility, an Android-based application was developed, allowing users to upload chest X-ray images and receive instant diagnostic results. The application successfully integrated the trained CNN model, offering a user-friendly interface suitable for healthcare professionals and patients alike. User testing demonstrated reliable performance, facilitating timely and accurate lung disease detection, particularly in areas with limited access to radiologists. These findings highlight the potential of CNNs in medical imaging and the critical role of mobile technology in expanding healthcare accessibility. This innovative approach not only improves diagnostic accuracy but also enables real-time disease detection, ultimately supporting clinical decision-making. Future research will focus on expanding the dataset, incorporating additional lung conditions, and optimizing the model for enhanced robustness in diverse clinical scenarios.
Co-Authors Anggawidia Wibaselppa Ary Maizary Aziz, RZ. Abdul Harry Tursulistyono Yani Achsan Joko Triloka M Said Hasibuan Mohd Zaki Zakaria Muhammad Daffa Arviano Putra Nicolas Melanie R Rizal Isnanto Sheeba Armoogum Sheeba Armoogum Sheeba Armoogum Tri Basuki Kurniawan Tri Basuki Kurniawan Tri Basuki Kurniawan Vinaye Armoogum Wahyuningdiah Trisari Harsanti Putri