UCF NLP Course; NLP Project 2019
Classifying Label
Cultural Harassment 1
Sexual Harassment 2
Personal Attacks 3
Labeling is done
Notes:
Some tweets were inexplicably split up in multiple lines. Folding them into one line shifted Maxim's starting point. Maxim's first change now sits at line 2957 . Given that it doesn't look like he had to collapse any tweets, and his first change occurred at line 2978 prior to me collapsing tweets, that suggests a current shift of -21 lines. So I assume Maxim's block starts at what is now line 2919.
Plan of Attack & Dates:
Data collection Feb 10 [V]
Data labeling into 3 categories Feb 17 [V]
Label | Raw Count | Percentage |
---|---|---|
Non-Bullying | 7,203 | 81.7% |
Cultural | 260 | 2.9% |
Sexual | 264 | 3.0% |
Personal | 1090 | 12.4% |
Data produced using the dataset_stats.py
script in the devtools
directory.
The Python program was ran under a 'virtualenv' built using
the requirements present in the repository's requirements.txt
file:
$ python3 -m virtualenv -p $(which python3) --always-copy ~/pythonenv/CAP6640_PROJ
...
$ . ~/pythonenv/CAP6640_PROJ/bin/activate
(CAP6640_PROJ)$ python3 -m pip install -r requirements.txt
...
(CAP6640_PROJ)$ python3 devtools/dataset_stats.py
Project proposal End of Feb [V]
The proposal, as submitted, is included in the project as the proposal.tex
and proposal.bib
files. Typesetting with LaTeX requires the files
acmcopyright.sty
and sig-alternate-05-2015.cls
.
Text tokenization Mar 5 [V]
The NLTK library has been added to the list of requirements for the project.
Text processing Mar 10 [V]
Classifier Apr 5 [:)]
SVM
BAYES
ANN