A social media and data mining project looking to identify and analyze wellness trends.
-
Download the
bigboy.csvdataset linked in the sources list. -
Install the requirements
pip install -r requirements.txt. -
Set up NLTK if you haven't already:
bash> python3 Python 3.12.0 ... >>> from nltk import download >>> download()
Click the "Download" button and wait for everything to download.
-
Run
processData.pyto generate the processed data CSVs. -
Run
getStatistics.pyto generate gra 5C06 phs, charts, and get stats.
data/- contains dataset (you should download the source data and put it here).img/- contains images (charts and graphs) rendered usingmatplotlibfrom the scripts.processed/- contains processed data.- You should use
relevant.csvas your data source for subsequent processing. This includes all columns from the original dataset, but only those rows that contain relevant hashtags. - You may also use
cleaned.csvas your data source for subsequent processing that needs all tweets. This just hase dates parsed in a format recognizable bypandas.
- You should use