Conquering vanishing gradient: Tensor Tree LSTM on aspect-sentiment classification

Our project focus on the problem of aspect specific sentiment analysis using recursive neural networks. Different from the previous studies where labels exist on every node of constituency tree, we have only one label each sentence, which is only on the root node, and it causes a severe vanishing gradient problem for both RNN and RNTN. To deal with such problem, we develop a classification algorithm specifically for data with only labels only for each sentence, instead of labels on each words. We deploy tree LSTM to solve the problem. and also reform it to a new model called ”Tensor Tree LSTM” as a combination of Tree LSTM and RNTN to conquer the vanishing gradient problem for RNTN.