Processing may take a few seconds...

Chapter


Title

Prototypical Convolutional Neural Network for a Phrase-Based Explanation of Sentiment Classification

Authors

[ 1 ] Instytut Informatyki, Wydział Informatyki i Telekomunikacji, Politechnika Poznańska | [ P ] employee

Scientific discipline (Law 2.0)

[2.3] Information and communication technology

Year of publication

2021

Chapter type

chapter in monograph / paper

Publication language

english

Keywords
EN
  • prototypes
  • explainable neural networks
  • text classification
  • phrase-based explanation
Abstract

EN The attention mechanisms are often used to support an interpretation of neural network based classification of texts by highlighting words to which the network paid attention while making a prediction. Following recent studies, the attention technique does not always provide a faithful explanation of the model. Thus, in this paper we study another idea of prototype-based neural networks. Although for texts they obtain promising results, they may provide explanations in the form of comparisons of whole (potentially long) documents or also run into problems with providing reliable explanations. To overcome it, in this work a new prototype-based convolutional neural architecture for text classification is introduced, which provides predictions’ explanations in the form of similarities to phrases from the training set. The experimental evaluation demonstrates that the proposed network obtains similar classification performance to the black-box convolutional networks while providing faithful explanations. Moreover, it is shown that a new method for dynamic tuning of the number of prototypes introduced in this paper offers performance gains against static tuning.

Date of online publication

01.01.2022

Pages (from - to)

457 - 472

DOI

10.1007/978-3-030-93736-2_35

URL

https://link.springer.com/chapter/10.1007/978-3-030-93736-2_35

Book

Machine Learning and Principles and Practice of Knowledge Discovery in Databases : International Workshops of ECML PKDD 2021, Virtual Event, September 13-17, 2021, Proceedings, Part I

Presented on

21th Joint European Conference on Machine Learning and Knowledge Discovery in Databases ECML PKDD 2021, 13-17.09.2021

Points of MNiSW / chapter

20.0

Points of MNiSW / conference (CORE)

140.0

This website uses cookies to remember the authenticated session of the user. For more information, read about Cookies and Privacy Policy.