NLP

Natural Language Processing

Dates: 2018 Jan 7 to 2018 Feb 4

in partnership with: Portland Data Science Group

Toxic Comments

Toxic comment data from Wulczyn, Ellery; Thain, Nithum; Dixon, Lucas (2016): Wikipedia Detox.

Warning: some comments in this data set contain offensive language and content. The project's goal is to flag for offensive behavior by predicting toxicity through automated NLP.

libraries used:

• NumPy, Pandas

• sklearn SciKit Learn

• nltk Natural Language Toolkit

• Matplotlib for visualization

• Keras for Deep Learning neural networks

Note 1: some, if not all, data files are found in this repository.

Note 2: All data may be found in this site: http://dive-into.info/ (Note: In my Windows Chrome browser, it was necessary to access this site incognito.)