Vigilancia Inteligente del Comercio Exterior: Detección de Anomalías en las Importaciones del Ecuador con Minería de Datos

Abstract

This study employs advanced data mining techniques to comprehensively analyze Ecuador's import data spanning three decades, from 1990 to 2021. By leveraging the k-means clustering algorithm, we meticulously identified anomalies within tariff items. The study utilized a comprehensive dataset that encompassed tariff items, years, and critical variables such as import volume, CIF and FOB values, import cost, and trade attractiveness. The data mining model generated insightful reports that accurately pinpointed an omalous patterns in tariff items. These reports empower experts to delve into the underlying causes of these anomalies, enabling them to make well-informed decisions to optimize Ecuador's import strategies. Our research underscores the transformative potential of data mining in detecting import anomalies, providing valuable intelligence for the strategic management of Ecuador's foreign trade. The findings contribute significantly to the prevention of customs fraud and unfair trade practices, ultimately enhancing the competitiveness of the country's import sector.