NLP-disaster-tweets

Classify tweets with #disaster hashtag into real disaster vs irrelevant tweets. The data set contains over 10,000 tweets, where roughly half are tweets about real disasters and half are irrelevant tweet.

Methods include:

1. Logistic regression with bag of words embedding
2. Logistic regression with word2vec embedding, incorporating the semantic meaning of each word
3. Decision tree
4. Random forest
5. Convolutional Neural Network, incorporating the text structure

Inspired by the blog https://blog.insightdatascience.com/how-to-solve-90-of-nlp-problems-a-step-by-step-guide-fda605278e4e