Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification

Fiachra Merwick; Y Bi; Peter Nicholl

doi:10.1007/978-3-030-82147-0_51

Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification

Fiachra Merwick
, Y Bi
, Peter Nicholl

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

Thispaperpresentsacomparativestudyontheperformanceofbinary- and multi-class Deep Neural Network classification models that have been trained with the optimized hyperparameters. Four sequential models are developed using Tensorflow and Keras to perform sentiment classification of product reviews posted on Amazon, IMDb, Twitter and Yelp. Preprocessing is carried out to cleanse text reviews and to extract informative features. 5-fold cross evaluation results demonstrate that the final multiclass model perform with an accuracy of 74.7% and 72.4%, whereas the binary model achieves of 80.0% and 69.5% indi- cating better performance. The conclusion of this study is that an optimized binary model architecture can be used to train multiclass classification models and save significant amounts of computing time.

Original language	English
Title of host publication	Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings
Subtitle of host publication	KSEM 2021
Pages	624–635
Number of pages	12
ISBN (Electronic)	978-3-030-82147-0
DOIs	https://doi.org/10.1007/978-3-030-82147-0_51
Publication status	Published online - 7 Aug 2021
Event	The 14th International Conference on Knowledge Science, Engineering and Management (KSEM 2021) - Tokyo, Japan Duration: 14 Aug 2021 → 16 Aug 2021 http://www.cloud-conf.net/ksem21/index.html

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	12816
ISSN (Print)	0302-9743

Conference

Conference	The 14th International Conference on Knowledge Science, Engineering and Management (KSEM 2021)
Abbreviated title	KSEM 2021
Country/Territory	Japan
City	Tokyo
Period	14/08/21 → 16/08/21
Internet address	http://www.cloud-conf.net/ksem21/index.html

Bibliographical note

Publisher Copyright:
© 2021, Springer Nature Switzerland AG.

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 3 Good Health and Well-being
SDG 16 Peace, Justice and Strong Institutions

Keywords

Sentiment analysis
Deep neural networks
Word embedding

Access to Document

10.1007/978-3-030-82147-0_51

Cite this

Merwick, F., Bi, Y., & Nicholl, P. (2021). Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification. In Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings: KSEM 2021 (pp. 624–635). (Lecture Notes in Computer Science; Vol. 12816). Advance online publication. https://doi.org/10.1007/978-3-030-82147-0_51

@inproceedings{c2a45afdb13948deba5482c2cefdeae8,

title = "Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification",

abstract = "Thispaperpresentsacomparativestudyontheperformanceofbinary- and multi-class Deep Neural Network classification models that have been trained with the optimized hyperparameters. Four sequential models are developed using Tensorflow and Keras to perform sentiment classification of product reviews posted on Amazon, IMDb, Twitter and Yelp. Preprocessing is carried out to cleanse text reviews and to extract informative features. 5-fold cross evaluation results demonstrate that the final multiclass model perform with an accuracy of 74.7\% and 72.4\%, whereas the binary model achieves of 80.0\% and 69.5\% indi- cating better performance. The conclusion of this study is that an optimized binary model architecture can be used to train multiclass classification models and save significant amounts of computing time.",

keywords = "Sentiment analysis, Deep neural networks, Word embedding",

author = "Fiachra Merwick and Y Bi and Peter Nicholl",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; The 14th International Conference on Knowledge Science, Engineering and Management (KSEM 2021), KSEM 2021 ; Conference date: 14-08-2021 Through 16-08-2021",

year = "2021",

month = aug,

day = "7",

doi = "10.1007/978-3-030-82147-0\_51",

language = "English",

isbn = "978-3-030-82146-3",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "624–635",

booktitle = "Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings",

url = "http://www.cloud-conf.net/ksem21/index.html",

}

Merwick, F, Bi, Y & Nicholl, P 2021, Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification. in Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings: KSEM 2021. Lecture Notes in Computer Science, vol. 12816, pp. 624–635, The 14th International Conference on Knowledge Science, Engineering and Management (KSEM 2021), Tokyo, Japan, 14/08/21. https://doi.org/10.1007/978-3-030-82147-0_51

Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification. / Merwick, Fiachra; Bi, Y ; Nicholl, Peter.
Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings: KSEM 2021. 2021. p. 624–635 (Lecture Notes in Computer Science; Vol. 12816).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification

AU - Merwick, Fiachra

AU - Bi, Y

AU - Nicholl, Peter

PY - 2021/8/7

Y1 - 2021/8/7

N2 - Thispaperpresentsacomparativestudyontheperformanceofbinary- and multi-class Deep Neural Network classification models that have been trained with the optimized hyperparameters. Four sequential models are developed using Tensorflow and Keras to perform sentiment classification of product reviews posted on Amazon, IMDb, Twitter and Yelp. Preprocessing is carried out to cleanse text reviews and to extract informative features. 5-fold cross evaluation results demonstrate that the final multiclass model perform with an accuracy of 74.7% and 72.4%, whereas the binary model achieves of 80.0% and 69.5% indi- cating better performance. The conclusion of this study is that an optimized binary model architecture can be used to train multiclass classification models and save significant amounts of computing time.

AB - Thispaperpresentsacomparativestudyontheperformanceofbinary- and multi-class Deep Neural Network classification models that have been trained with the optimized hyperparameters. Four sequential models are developed using Tensorflow and Keras to perform sentiment classification of product reviews posted on Amazon, IMDb, Twitter and Yelp. Preprocessing is carried out to cleanse text reviews and to extract informative features. 5-fold cross evaluation results demonstrate that the final multiclass model perform with an accuracy of 74.7% and 72.4%, whereas the binary model achieves of 80.0% and 69.5% indi- cating better performance. The conclusion of this study is that an optimized binary model architecture can be used to train multiclass classification models and save significant amounts of computing time.

KW - Sentiment analysis

KW - Deep neural networks

KW - Word embedding

UR - https://www.scopus.com/pages/publications/85113738303

U2 - 10.1007/978-3-030-82147-0_51

DO - 10.1007/978-3-030-82147-0_51

M3 - Conference contribution

SN - 978-3-030-82146-3

T3 - Lecture Notes in Computer Science

SP - 624

EP - 635

BT - Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings

T2 - The 14th International Conference on Knowledge Science, Engineering and Management (KSEM 2021)

Y2 - 14 August 2021 through 16 August 2021

ER -

Merwick F, Bi Y , Nicholl P. Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification. In Knowledge Science, Engineering and Management - 14th International Conference, KSEM 2021, Proceedings: KSEM 2021. 2021. p. 624–635. (Lecture Notes in Computer Science). Epub 2021 Aug 7. doi: 10.1007/978-3-030-82147-0_51

Performance Evaluation of Multi-class Sentiment Classification Using Deep Neural Network Models Optimised for Binary Classification

Abstract

Publication series

Conference

Bibliographical note

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this