Resumen |
This paper shows an application of data science in the healthcare system by using the Social Inclusion Indicators (ISS) of each entity in Mexico for 25 years to make a clustering based on the lack of primary healthcare. Multiple procedures were applied, like cleaning and transformation of open data published by the Mexican Health Department, the imputation of missing values. With the complete information, data was scaled, and then one of the most common clustering algorithms was applied, which is K-Means. This algorithm was initialized with previously defined centroids to make it more standardized and make it easier to notice changes amongst the classes through the years. Six clusters were defined using previous works. All the implementations were made in Python using the Scikit-Learn library to apply the algorithms and measure performance, like K-Means and Mean Squared Error respectively. Results obtained were displayed using Tableau to observe in a more interactive way, how the classes had changed over the years. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
|