Implementation of Data Science Techniques in the ACM Computing Classification System

In 2012, ACM defined a Computing Classification System (CCS), which substitutes an older version developed in 1998. This CCS relies on a semantic vocabulary as the single source of categories and concepts that reflect the state of the art of the computing discipline. This paper shows the application to the CCS of two Data Science techniques used for classification tasks: Naive Bayes (NB) and Convolutional Neural Network (CNN). The project comprises two tasks: developing a framework with the general roles and data structure and applying NB and CNN to CCS classification. After the adjustment of some model parameters, the results are promising. The classification with NB provides an accuracy of 74%, while the accuracy with CNN was 72%.