EXTRACTION DES CONNAISSANCES À PARTIR DES DONNÉES FIREBASE POUR L’ANALYSE ÉPIDÉMIOLOGIQUE BASÉ SUR LE CLUSTERING DBSCAN.

Abstract

A digital shift toward precision epidemiology is required due to the quick evolution of health emergencies. In the context of epidemiological monitoring within the Provincial Coordination of PNMLS Kwilu, this study introduces a novel method for knowledge extraction from large datasets stored on the Firebase NoSQL platform. Converting diverse data sources into tools for strategic decision-making is the goal. We developed an analytical pipeline that integrated the DBSCAN (Density-Based Spatial Clustering of Applications with Noise) spatial clustering algorithm in order to accomplish this. In contrast to conventional techniques, DBSCAN found contagion hotspots with irregular shapes while eliminating outliers or solitary cases that were regarded as statistical "noise." The findings show that dynamic mapping of high-prevalence areas is made possible by the combination of Firebase's velocity and DBSCAN's segmentation power. This study optimizes the distribution of healthcare resources throughout the province by giving PNMLS Kwilu decision-makers a targeted intervention capacity.