Tagtwitter

• 
  Feb 19, 2018CompSciMigrated from old WordPress site

  Twitter Sentiment Analysis using combined LSTM-CNN Models

  Combining CNNs and LSTMs for tweet sentiment analysis — the intuition behind each network, two stacked architectures (CNN-LSTM vs. LSTM-CNN), and why ordering the LSTM first wins, trained on 1.5M+ labeled tweets.

  Read the post
• 
  Dec 20, 2016CompSciMigrated from old WordPress site

  Twitter Sentiment Analysis with Neural Networks

  Hand-coding a Feed-Forward Neural Network from scratch to classify tweets as positive or negative, comparing Bayesian probabilities, Keras word embeddings, and a hand-built feature vector (~70% accuracy) against a Keras NN.

  Read the post