Prohibited Comments Classification

José Lise

José Lise

Data Science Portfolio

Prohibited Comments Classification

less than 1 minute read

This work is inspired by an exercise from the course NLP Course. The goal is to build a model to ban offensive social media comments. The dataset is rather small (it contains about 3500 comments). For this study, we will:

Build from scratch a Naive Bayes classifier using Bag of Words features
Compare the Naive Bayes classifier with a logistic regression model
Compare logistic regression model with bag of words features with TF-IDF features
Use word vectors features

The details are available in the linked Jupyter Notebook.

Ban

Share on

Twitter Facebook LinkedIn

You may also enjoy

Neural machine translation with attention

16 minute read

NMT, NLP, Machine Learning, Python, Text, Deep Learning, GRU, translation, RNN, Attention, Tensorflow

Neural characters language models

15 minute read

Language Model, NLP, Machine Learning, Python, Text, Deep Learning, CNN, RNN, LSTM, Convolution

N-gram language models or how to write scientific papers

10 minute read

Language Model, NLP, Machine Learning, Python, Text

Multi-class Multi-label Classification Model based on the ONET Job Database

less than 1 minute read

Classification, Deep Learning, Keras, Tensorflow, Sk-learn, Neural Network, NLP, Machine Learning, Python, Multi-class, Multi-label