Full Program »
Detection of Taxpayers with High Probability of Non-payment.–An Implementation of a Data Mining Framework
Due to limitations in tax administrations, such as, staff, tools, time, etc., tax administrations seek to recover debts in the early stages of control, where the cost of collection is lower than in the subsequent stages. In order to optimize the debt management process and contribute to decision-making, this work proposes a framework using deep learning techniques to predict debts of taxpayers with high probability of non-payment. A group of debts of a tax administration was used to generate the model to estimate the risk of non-payment. The performance obtained was 90%. A Concordance index metric was used to measure the performance.