GitHub - arlobryer/insult_detect: Naive Bayes Classifier/Maxent implemented using the nltk module and libraries to assign a probability on whether a comment in a thread is an insult.

Branches Tags

Name		Name	Last commit message	Last commit date
Latest commit History 31 Commits
data		data
.gitignore		.gitignore
README		README
classify.py		classify.py
dat_read.py		dat_read.py
metrics.py		metrics.py
run.py		run.py
text_extract.py		text_extract.py

Repository files navigation

An attempt at a classifer for insulting comments in social commentary.

Details and datasets are found at:
http://www.kaggle.com/c/detecting-insults-in-social-commentary