Thank you @crokkon for the reply :). Here is some additional information about the data collection process. The data was collected from the Steemit page from the Tags section on Thursday, February 8, 2018 at around 10 am CET. The scraped data (the tags and associated counts and measurements: posts, comments, payouts) was imported and processed using Python while for the NLP processing the wordnet library for Python was employed. There was no reason in particular for selecting the data on this time or day, but the need of accessing real data for performing the analysis and drawing the conclusions.