Subsentence Extraction from Text Using Coverage-Based Deep Learning Language Models

JongYoon Lim,Bruce MacDonald,Sanghyub John Lee,Ho Seok Ahn,Norina Gasteiger,Inkyu Sa

doi:10.3390/s21082712

Abstract

Sentiment prediction remains a challenging and unresolved task in various research fields, including psychology, neuroscience, and computer science. This stems from its high degree of subjectivity and limited input sources that can effectively capture the actual sentiment. This can be even more challenging with only text-based input. Meanwhile, the rise of deep learning and an unprecedented large volume of data have paved the way for artificial intelligence to perform impressively accurate predictions or even human-level reasoning. Drawing inspiration from this, we propose a coverage-based sentiment and subsentence extraction system that estimates a span of input text and recursively feeds this information back to the networks. The predicted subsentence consists of auxiliary information expressing a sentiment. This is an important building block for enabling vivid and epic sentiment delivery (within the scope of this paper) and for other natural language processing tasks such as text summarisation and Q&A. Our approach outperforms the state-of-the-art approaches by a large margin in subsentence prediction (i.e., Average Jaccard scores from 0.72 to 0.89). For the evaluation, we designed rigorous experiments consisting of 24 ablation studies. Finally, our learned lessons are returned to the community by sharing software packages and a public dataset that can reproduce the results presented in this paper.

Highlights

Understanding human emotion or sentiment is one of the most complex and active research areas in physiology, neuroscience, neurobiology, and computer science
Evaluation metrics used for evaluating the model performances (Jaccard score, AUC, and F1 scores) are defined, followed by classification and coverage-based subsentence extraction models
[TR_CORR]_[SC]_ROB] can be interpreted by the RoBERTa-base model trained on the corrected dataset for the sentiment classification task

Summary

Introduction

Understanding human emotion or sentiment is one of the most complex and active research areas in physiology, neuroscience, neurobiology, and computer science. Sentiment retrieval from personal notes or extracting emotions in multimedia dialogues generates a better understanding of human behaviours (e.g., in crime or abusive chat). Sentiment analysis and emotion understanding have been applied to the field of Human–Robot Interaction (HRI) in order to develop robotics that can form longer-term relationships and rapport with human users by responding with empathy. This is crucial for maintaining interest in engagement when the novelty wears off. Robots that can identify human sentiment are more personalized, as they can adapt their behaviour or speech according to the sentiment

Objectives

Results

Discussion

Conclusion