Modelo predictivo para determinar el fracaso de matemáticas en grado 11 usando Machine Learning

Abstract

The predictive model to determine eleventh grade math failure using machine learning is the result of a data analysis at the Almirante Padilla de Usme Public School. This model is made up of a process of data integration, variable description, data cleaning, variable elimination, correlation analysis, data transformation and data balancing. These modules were implemented under the Python programming language to build various applications intelligent companies that use machine learning and for the analysis process, the CRISP-DM methodology (Cross Industry Standard Process for Data Mining) was chosen, it covers the phases of a project, their respective tasks, and the relationships between these tasks contemplates the process of data analysis as a professional project, thus establishing a much richer context that influences modeling.