The final project for analyzing disater-related twitter
Tweets were collected using the codes in here https://github.com/DisasterMasters
The jupyter notebook LDA_Topic_Modeling.ipynb contains the codes for LDA topic modeling.
TwitterAnalysisFull.ipynb contains the full Python code used for cleaning, separating, and applying K-Means and LDA to the dataset.
Twitter_DisasterAnalysis_finalReport.pdf contains the final report of this project.
CrisisLexT6/T26 labeled datasets:
https://crisislex.org/data-collections.html#CrisisLexT6
You will find more datasets here (including a newly released multi-modal text/image dataset - resource 5):
Here you will find a site where crawling is done with one click:
http://aidr-web.qcri.org/AIDRFetchManager/public.jsp
This is the TwitterNLP software for entity tagging:
https://github.com/aritter/twitter_nlp
This is code for text classification using CNN-RNN (either LSTM or GRU):
https://github.com/jiegzhan/multi-class-text-classification-cnn-rnn
This is the website for GloVe: