Abstract

The Sistema de Medición de la Calidad de la Educación(Education Quality Measurement System, SIMCE) is anannual survey designed to evaluate the Chilean educa-tional system through standard tests that measure theabilities and knowledge of the students. The results ofthe SIMCE exams provide important information for an-alyzing the learning processes of the students. Addition-ally, these results allow the identification of strengths andweaknesses for the elaboration of public policies.In this paper, we design a semi-supervised regressorcalled TTOSOM as Regression Model (TTOSOM-RM),that inherits the properties of the neural network calledTree-based Topology Oriented SOM (TTOSOM) [1]. Thegoal is to predict the performance of fourth grade studentsof the Chilean educational system in the SIMCE test.The proposed model successfully predicts the SIMCEscores, producing a lower absolute mean error when com-pared with other state-of-the-art methods.