Stance detection has been defined as automatically detecting whether the author of apiece of text is in favor of the given target or against it. [1]
during my 2.5 month internship at icog-labs i developed a neural network architecture for stance detection
the neural network have 2 self attention layers relating different positions of a single sequence in order to compute a representation of the same sequence [2] body and head representations
pass through the self-attention layers the output of the attention representation will be concatenated and feed it to a fully-connected layer
This is also a part of fake news detection competition
wget http://nlp.stanford.edu/data/glove.6B.zip
pip install -r requirements.txtrun ./notebook/Stance Detaction.ipynb
docker build -t stance .
docker run -p 8888:8888 stance