Dialogue State Tracking using Long Short Term Memory Neural Networks

We propose a dialogue state tracker based on long short term memory (LSTM) neural networks. LSTM is an extension of a recurrent neural network (RNN), which can better consider distant dependencies in sequential input. We construct a LSTM network that receives utterances of dialogue participants as input, and outputs the dialogue state of the current utterance. The input utterances are separated into vectors of words with their orders, which are further converted to word embeddings to avoid sparsity problems. In experiments, we combined this system with the baseline system of the dialogue state tracking challenge (DSTC), and achieved improved dialogue state tracking accuracy.