A Web-Based Vector Reporting Information System Using Decision Trees for Risk Classification (Case Study: Manado Class 1 Health Quarantine Office, Manado Seaport Working Area)

Vector-borne diseases remain a critical public health challenge, particularly in tropical port cities where international maritime traffic increases the risk of introducing infected vectors. At the Class 1 Health Quarantine Center of Manado (BKKK Manado), traditional paper-based vector reporting workflows have caused delays, transcription errors, and inconsistent risk assessments, hindering timely and evidence-based decision-making. This study aims to develop and evaluate a web-based vector reporting information system integrated with a C4.5 Decision Tree classifier to automate risk classification and improve operational efficiency. An applied research approach using a research-and-development (R&D) methodology was employed, involving system design, implementation, and empirical evaluation at the Manado seaport. Data were collected from 312 historical vector surveillance records, field observations, and officer interviews. System performance was assessed through classification accuracy, functional testing, usability evaluation (System Usability Scale), and a time-efficiency comparison with paper-based reporting. The resulting system achieved 92.1% classification accuracy, a macro-averaged F1-score of 0.91, a 100% functional test pass rate, and an 80.7% reduction in reporting time, while usability was rated “Excellent” by officers. The study concludes that the web-based system effectively enhances vector surveillance and decision-making. Future research should focus on expanding datasets, integrating with national health platforms, and exploring alternative classifiers to improve scalability and robustness for broader vector-borne disease monitoring.