This task should be done on a dataset from the stackexchange database dump, any dataset will do.

Create a single Jupyter/IPython notebook (see the Artefacts section below for all the requirements), where you perform what follows. 1. Convert all the data tables (Badges, Comments, PostHistory, PostLinks, Posts, Tags, Users, Votes) from XML to CSV, using custom code that you write yourself. Ideally, you should write a Python function that takes a single input file name (.xml) and output file name (.csv) and performs the conversion of a single dataset. 2. Load the CSV files as pandas data frames. 3. Create at least five nontrivial data visualisations and/or tables, at least three of which are based on the extraction of information from text (e.g., tags, keywords, locations, etc.). 4. Draw insightful and interesting conclusions. Do not forget to reflect on the potential data privacy and ethics issues that arise during the data analysis process. The PDF version of the report must be at least 10 pages long (not including the data conversion/import part). Make it aesthetic and interesting to read