First, research$/$gather the data:

$1.$ ChooseoneStackExchangesitedealingwithtopicsthatyoufindinteresting;seehttps:$//$stackexc hange.com$/$sites$?$view$=$list#traffic for a list. The site cannot be too small, but also avoid selecting any of the largest ones $($especially StackOverflow, Mathematics$)$ unless you really want to challenge yourself. As a rule of thumb, let$$s say that the site must have at least $10,000$ questions and $10,000$ answers.

This document was originally developed by Dr$.$ Marek Gagolewski. It was subsequently revised by Dr$.$ Yang Li $($Kelvin$)$ during the work at the School of Information Technology, Deakin University, for the unit SIT$220/731$ Data Wrangling, Trimester $1,2024.$

$2.$ Downloadthesite$$smostrecentdatadumpfromhttps:$//$archive$.$org$/$details$/$stackexchange$/.$

$3.$ Readthedescriptionofallthedatatablespublishedathttps:$//$meta$.$stackexchange.com$/$questio

ns$/2677/.$

Then, create a single Quarto $.$qmd file$1$ that you will be rendering to a PDF report $($how to do that you will

have to learn yourself $$ this is part of this HD$-$level task$),$ where you perform what follows.

$1.$ Convertallthedatatables$($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.

$2.$ LoadtheCSVfilesaspandasdataframes.

$3.$ Createatleastfivenontrivialdatavisualisationsand$/$ortables$,$atleastthreeofwhicharebasedon the extraction of information from text $($e$.$g$.,$ tags, keywords, locations, etc.$).$ You must demon$-$ strate that you have learned how to write your own regular expressions $($regexes$).$

$4.$ Drawinsightfulandinterestingconclusions.Donotforgettoreflectonthepotentialdataprivacy and ethics issues that arise during the data analysis process.

This HD$-$level task is purposely under$-$defined $$ you will not be told precisely what to do$.$ Your aim is to generate some interesting insights into data featuring lots of textual information.

In the course of the report preparation, you should apply a wide range of data frame wrangling and text processing techniques. In particular, you must demonstrate that you mastered regular expressions.

Do not use pie charts $($as we discussed during the lecture$).$ Go beyond the basic plots that we have covered in this course. Draw at least one map $($e$.$g$.,$ of the world$)$ and a word cloud.