Output¶

The first step followed consists in convert the input PDFs in XML files using Grobid, in this way, we can analyze them from python.

For the Draw a keyword cloud based on the abstract information objective, we get the abstract of each article by finding the "abstract" tag in the XML. Then, we create a wordcloud image of the abstract using the Wordcloud library.

For the Create a visualization showing the number of figures per article objective, we count the number of figures of each article by finding the "figure" tag in the XML. Then, we create an histogram image showing the number of figures per article.

For the Create a list of the links found in each paper objective, we get all the links that appear in the XML text by finding text elements complying with the regular expression that stays only with texts starting with "https//". Then, the links are saved in a text file in which each line is a link.