Mapeo y síntesis a través de esquemas de clasificación de un dataset referente a artículos científicos sobre el coronavirus a través de lista y nube de palabras de los artículos (etapa 2)

Abstract

The optimization attempt of a Systematic Mapping Study (SMS) through Machine Learning has two stages, in the first stage of this study that took the name of "Classification and Mapping of a dataset of Scientific Articles on SARS-CoV2 through List and Cloud of Words”, the author of the work Moromenacho shows the stages of: Definition of the investigation, Search of articles, keywords and Selection of articles, in which the Python and Excel tools are involved, for the automation the Dataset was used. of CORD-19 that was downloaded from Kaggle, in the implementation a file with extension .xlsx was obtained where titles, abstracts and most used words, total words of the text of the articles are found. For this second one, a data validation is carried out, we will deal with the stages of: Selection of relevant articles, keyword search, mapping process and data extraction, the tools used with Python were Apache Spark and the Non-Negative Factorization method of Matrices (NMF) of Machine Learning, to achieve this classification it was necessary to start from the inclusion and exclusion criteria, generate keywords through the model and finally as the main objective to be able to see towards which social aspect a trend is being generated.