Jurnal Aplikasi Sains Data

Dijkstra's Algorithm for Optimizing Humanitarian Aid Distribution Routes to Flood Victims in Cerme District, Gresik

Thu, 30 Apr 2026 00:00:00 +0000

This study presents the development and analysis of a system designed to optimize the distribution routes of social aid during flood emergencies in the Cerme District, Gresik Regency. The primary objective is to ensure that logistical operations, particularly the delivery of aid to affected villages, are carried out in the most efficient and timely manner. To achieve this, Dijkstra’s Algorithm is employed due to its well-established reliability in computing the shortest path between nodes in a weighted graph. The graph used in this research is constructed based on real-world spatial data, with each node representing a village and the edges representing actual road distances obtained from mapping services. The system is implemented using an Object-Oriented Programming (OOP) paradigm in Python, which ensures modularity and scalability of the codebase. For graph modeling and shortest path computation, the NetworkX library is utilized, while the graphical user interface (GUI) is built using Tkinter to provide an interactive and user-friendly experience. The application enables users to select starting and destination points from dropdown menus, compute the shortest route dynamically, and visualize it on an interactive graph complete with route details and distances. Experimental trials were conducted by simulating various flood scenarios, and the results demonstrated that the system successfully identified optimal aid routes with minimized travel distances. These outcomes confirm the practicality and effectiveness of the proposed method. Moreover, the ability to update the graph dynamically allows the system to adapt to changes in road accessibility due to flooding. This makes the tool highly applicable in real-world disaster response scenarios. In conclusion, the developed application offers a valuable solution for both local government agencies and humanitarian volunteers, helping to improve coordination, reduce delivery time, and ensure that aid reaches flood-affected communities as efficiently as possible.

Comparative Analysis of Hierarchical Clustering and K-Medoids for Clustering Cases of Childhood Respiratory Diseases in Lamongan Regency

Thu, 30 Apr 2026 00:00:00 +0000

Abstract— Respiratory diseases affecting children remain a significant health issue in Indonesia, including in Lamongan Regency. The region faces challenges related to pediatric respiratory illnesses, particularly Childhood Tuberculosis, Pneumonia in toddlers, and Cough in toddlers, which impact children's quality of life and development. Therefore, understanding the spatial distribution and correlation patterns among these diseases is essential to support more targeted health intervention planning. This study analyzes the distribution patterns of pediatric respiratory diseases in Lamongan Regency and clusters regions based on similarities in the number of cases using an unsupervised learning approach. The method employed is Hierarchical Clustering with four distance calculation techniques: single, complete, average, and ward linkage and K-Medoids with two distance calculation techniques: euclidean and manhattan distance. The data, sourced from the Lamongan District Health Office, include four numerical variables related to respiratory diseases, aggregated by sub-districts. Data normalization was carried out using standardization, and cluster quality was evaluated using three internal metrics: Silhouette Score, Davies-Bouldin Index (DBI), and Calinski-Harabasz Index (CHI). The analysis results indicate that the optimal number of clusters is three. Among all methods tested, the Hierarchical Clustering with ward linkage method yielded the best performance, with a Silhouette Score of 0.5447, a DBI of 0.5884, and a CHI of 20.3018. These results demonstrate that the ward linkage method is the most effective in clustering regions based on the characteristics of pediatric respiratory disease cases and can be used for mapping priority health intervention areas in Lamongan Regency.

A Quantitative Analysis of Economic Strategy and Its Influence on Final Ranking in Magic Chess Game Using Machine Learning

Thu, 30 Apr 2026 00:00:00 +0000

Economic management is a fundamental strategic pillar in auto-battler games such as Magic Chess, but its quantitative impact on player performance has not been extensively studied. This research aims to empirically measure the predictive ability of economic variables on players' final rankings. We analyzed a dataset consisting of 57 match records from players at the ‘Grandmaster’ ranking level. Two modeling approaches, Multiple Linear Regression and Random Forest, were used to predict players' final rankings (values 1–8) based on three primary economic features: total gold spent, re-roll frequency, and average economic bonus. The results from the Linear Regression model showed a Mean Squared Error (MSE) of 0.5496. However, the most significant finding was the R-squared value, which was only 0.016. This extremely low R-squared value indicates that the economic variables analyzed could only explain 1.6% of the total variance in players' final rankings. The conclusion of this study is that economic metrics alone are insufficient to build a reliable model for accurately predicting final rankings. This strongly suggests that other strategic factors, such as synergy composition, item allocation, and tactical decisions on the game board, have a far more dominant influence in determining a player's success in high-level Magic Chess.

Spatial Modeling of Factors Determining Active Family Planning Participation in East Java: A Geographically Weighted Regression and Elastic Net Approach

Ardiana Fatma Dewi, I Nyoman Kresna Wira Yudha, Muhammad Nasrudin — Thu, 30 Apr 2026 00:00:00 +0000

This research identifies regional variations and determinants of active family planning (KB) participation in East Java using spatial modeling. The study utilizes data from the 2024 Family Information System (SIGA) of BKKBN East Java. To address multicollinearity and high-dimensional data, the Elastic Net method—combining Ridge and Lasso penalties—was employed for variable selection, retaining 6 out of 10 initial variables. Global modeling through Ordinary Least Squares (OLS) showed an Adjusted of 0.668. However, a Moran’s I test on the residuals revealed significant spatial autocorrelation (Z-score = 2.5677, p = 0.0102), justifying the use of Geographically Weighted Regression (GWR). The GWR model, using a Fixed Gaussian kernel with a bandwidth of 103.63, improved performance with an Adjusted of 0.7348. The results demonstrate spatial heterogeneity, where factors such as unmet need, households with children, and welfare levels have varying impacts across different districts. This spatial visualization helps identify priority areas for strategic resource allocation to enhance KB program efficiency