Abstract A tool using machine learning which organizes, clean and analyzes data in massive quantities was developed. So that the user can make predictions about different aspects of a list of more than 2000 students, using the k-nearest neighbor machine learning method. Data were obtained from a Colombian university, which has previously it has been cleaned and organized to accommodate the input parameters into a script. The computational tool was written in python from Jupyter notebook. The script is able to perform analysis predictive of the different filters previously chosen for decision raking by the example of marketing, the tool will keep track of financial actions and of the different categories chosen for students at the university, allowing a global analysis and thus choose the best options from all data granted. Among these categories we can classify method of payment, the value paid, what is the method of most commonly used payout by students, individual averages per semester and academic programs, addresses, and academic careers among others.