Deakin University
Browse

File(s) under permanent embargo

A Mask-Based Logic Rules Dissemination Method for Sentiment Classifiers

Disseminating and incorporating logic rules inspired by domain knowledge in Deep Neural Networks (DNNs) is desirable to make their output causally interpretable, reduce data dependence, and provide some human supervision during training to prevent undesirable outputs. Several methods have been proposed for that purpose but performing end-to-end training while keeping the DNNs informed about logical constraints remains a challenging task. In this paper, we propose a novel method to disseminate logic rules in DNNs for Sentence-level Binary Sentiment Classification. In particular, we couple a Rule-Mask Mechanism with a DNN model which given an input sequence predicts a vector containing binary values corresponding to each token that captures if applicable a linguistically motivated logic rule on the input sequence. We compare our method with a number of state-of-the-art baselines and demonstrate its effectiveness. We also release a new Twitter-based dataset specifically constructed to test logic rule dissemination methods and propose a new heuristic approach to provide automatic high-quality labels for the dataset.

History

Volume

1

Pagination

394-408

Location

Dublin, Ireland

Start date

2023-04-02

End date

2023-04-06

ISSN

0302-9743

eISSN

1611-3349

ISBN-13

9783031282430

Language

eng

Publication classification

E1 Full written paper - refereed

Editor/Contributor(s)

Goos G, Hartmanis J

Title of proceedings

ECIR 2023 : Proceedings of the 45th European Conference on Information Retrieval

Event

Information Retrieval. Conference (2023 : 45th : Dublin, Ireland)

Publisher

Springer Nature Switzerland

Place of publication

Lausanne, Switzerland

Series

Advances in Information Retrieval

Usage metrics

    Research Publications

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC