PolarityMining-StackOverflow

This is the code repository of my seminar paper "Polarity mining of Stack Overflow answers in the C and Python subcommunities". It contains all my Python and SQL scripts for data processing, data analysis and the final presentation

DataProcessing files:

Files	Usage
analyzer.py	Used to read the data from the csv data dumps, input it into VADER and then inserts everything into the PostgresSQL database
codecleaner.py	Used to clean HTML tags from the answer bodies and remove code and links completely
csvjointer.py	Used to join the csv files from the monthly data dumps
statreader.py	Used to generate some statistics about the data dumps before they were processed
tagprinter.py	Used to print example usages of HTML tags in the answer body
tagreader.py	Used to find all tags that occur withing the data dumps

Data Analysis files:

Files	Usage
answer_number_gen.py	Used to generate random numbers to get a set of numbers for the manual rating process
correlationCalculator.py	Used to calculate the pearson correlation of 2 data dumps
Edited_Answer_Code.py	Used to calculate statistics about answers which were edited and not edited from data dumps
manual_answer_check_puller.py	Used to process data from the database for the manual rating process
manual_rater.py	Used to manually rate answers from data dumps in the console
presentation.py	Used to manually rate answers during the presentation through a graphical interface instead of the console
results_of_manual_rating.py	Uses to display the results of the manual rating in a nice way for the paper
Visualizer.py	Used to generate LateX graphs with Matplotlib from data dumps
SQL files	All the database queries used to gain knowledge from the data. Most queries are commented and explained.

Data Acquision files:

Files	Usage
QueryFile.sql	Used to get the data from the live database of Stack Exchange Meta. Only the important skeleton queries, because one query for a 3 month interval is the same as for all other months with only changes to the time interval