%0 Conference Proceedings
%T Explaining Recurrent Neural Network Predictions in Sentiment Analysis
%A Arras, Leila
%A Montavon, Grégoire
%A Müller, Klaus-Robert
%A Samek, Wojciech
%Y Balahur, Alexandra
%Y Mohammad, Saif M.
%Y van der Goot, Erik
%S Proceedings of the 8th Workshop on Computational Approaches to Subjectivity, Sentiment and Social Media Analysis
%D 2017
%8 September
%I Association for Computational Linguistics
%C Copenhagen, Denmark
%F arras-etal-2017-explaining
%X Recently, a technique called Layer-wise Relevance Propagation (LRP) was shown to deliver insightful explanations in the form of input space relevances for understanding feed-forward neural network classification decisions. In the present work, we extend the usage of LRP to recurrent neural networks. We propose a specific propagation rule applicable to multiplicative connections as they arise in recurrent network architectures such as LSTMs and GRUs. We apply our technique to a word-based bi-directional LSTM model on a five-class sentiment prediction task, and evaluate the resulting LRP relevances both qualitatively and quantitatively, obtaining better results than a gradient-based related method which was used in previous work.
%R 10.18653/v1/W17-5221
%U https://aclanthology.org/W17-5221
%U https://doi.org/10.18653/v1/W17-5221
%P 159-168