Machine learning for libraries with Python libraries

practical case in the Library of Congress of Chile

Abstract

This document focuses on the area of machine learning from data, applied to internal processes of a library. This is practical work associated with the development of an application in Python that uses libraries developed for automated learning work. An unsupervised data analysis methodology was applied, called the K-means method (K-medias in Spanish), which allows the data to be segmented or classified into groups to extract common characteristics. Data associated with the library collections were used. The developed code is shared, and visualisations of the data are shown.