Text Classification using Triplet Capsule Networks

Wu, Yujia; Li, Jing; Chen, Vincent; Chang, Jun; Ding, Zhifeng; Wang, Zhi

doi:10.1109/ijcnn48605.2020.9207201

Cited by 11 publications

(2 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since spatial relations can represent crucial properties depending on the concrete task, it should be quite natural that capsules are forced to learn this structural context. In coherence with this, Wu et al [17] stated that capsules could depict grammatical structure information and spatial distance of local features.…”

Section: ) Representation Learningmentioning

confidence: 88%

“…Chen et al [19] introduced a transfer capsule network for learning sentiment polarity in diverse aspects based on existing text classification models. Wu et al [17] proposed a triplet capsule network with an adjusted triplet loss function to force the network's ability to assign latent features with low discriminative power to the correct class and, therefore, improve classification effectiveness. Xiao et al [20] designed a multi-task learning CapsNet which involves a task routing procedure to cluster shared features by their relevance for different tasks.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Next-Generation Neural Networks: Capsule Networks With Routing-by-Agreement for Text Classification

Steur

Schwenker

2021

IEEE Access

View full text Add to dashboard Cite

These days, neural networks constantly prove their high capacity for nearly every application case and are considered as key technology for learning systems. However, neural networks need to continuously evolve for managing new arising challenges like increasing task complexity, explainability of decision making processes, expanded problem domains, providing resilient and robust systems etc. One possible enhancement of traditional neural networks constitutes the innovative Capsule Network (CapsNet) technology, which combines the expressiveness of distributed entity representations with an intelligent and interpretable signal propagation, named as routing-by-agreement. Since CapsNets represent a relatively young acquirement, further research is essential for gaining profound knowledge about CapsNet theory and best practices for diverse application areas. This paper wants to contribute to the progress of CapsNets for the task of text classification. For this purpose, various research questions about this technology get formulated and experimentally answered with the aid of six selected datasets. In addition, this paper serves as a possible starting point for researchers as well as for practitioners to deal with CapsNets in the text domain, by supplying a survey about its theory, text classification basics and the combination of both areas. The analysis results empirically prove the robustness of CapsNets with routing-by-agreement for a wide spectrum of net architectures, datasets and text classification tasks. Hence, CapsNets can be viewed as a next-generation neural network technology, which offers high potential as text classification method and should be topic of future research.

show abstract

Section: ) Representation Learningmentioning

confidence: 88%