Investigating Capsule Networks with Dynamic Routing for Text Classification

In this study, we explore capsule networks with dynamic routing for text classification. We propose three strategies to stabilize the dynamic routing process to alleviate the disturbance of some noise capsules which may contain "background" information or have not been successfully trained. A series of experiments are conducted with capsule networks on six text classification benchmarks. Capsule networks achieve state of the art on 4 out of 6 datasets, which shows the effectiveness of capsule networks for text classification. We additionally show that capsule networks exhibit significant improvement when transfer single-label to multi-label text classification over strong baseline methods. To the best of our knowledge, this is the first work that capsule networks have been empirically investigated for text modeling.

PDF Abstract EMNLP 2018 PDF EMNLP 2018 Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Text Classification AG News Capsule-B Error 7.4 # 12
Sentiment Analysis CR Capsule-B Accuracy 85.1 # 7
Sentiment Analysis MR Capsule-B Accuracy 82.3 # 6
Sentiment Analysis SST-2 Binary classification Capsule-B Accuracy 86.8 # 64
Subjectivity Analysis SUBJ Capsule-B Accuracy 93.8 # 11
Text Classification TREC-6 Capsule-B Error 7.2 # 14

Methods


No methods listed for this paper. Add relevant methods here