Siamese neural network is a class of neural network architectures that contain two or more identical subnetworks. identical here means they have the same configuration with the same parameters and weights. Parameter updating is mirrored across both subnetworks. Using MaLSTM model(Siamese networks + LSTM with Manhattan distance) to detect semantic similarity between Tweet pairs.
Below is the architecture description for the same.
http://www.mit.edu/%7Ejonasm/info/MuellerThyagarajan_AAAI16.pdf https://medium.com/mlreview/implementing-malstm-on-kaggles-quora-question-pairs-competition-8b31b0b16a07
https://drive.google.com/file/d/0B7XkCwpI5KDYNlNUTTlSS21pQmM/edit?usp=sharing