Performance Analysis of Hybrid Architectures of Deep Learning for Indonesian Sentiment Analysis

Theresia Gowandi, Hendri Murfi, Siti Nurrohmah

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Sentiment analysis is one of the fields of Natural Language Processing that builds a system to recognize and extract opinions in the form of text into positive or negative sentiment. Nowadays, many researchers have developed methods that yield the best accuracy in performing analysis sentiment. Three particular models are Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM), and Gated Recurrent Unit (GRU), which have deep learning architectures. CNN is used because of its ability to extract essential features from each sentence fragment, while LSTM and GRU are used because of their ability to memorize prior inputs. GRU has a more straightforward and more practical structure compared to LSTM. These models have been combined into hybrid architectures of LSTM-CNN, CNN-LSTM, and CNN-GRU. In this paper, we analyze the performance of the hybrid architectures for Indonesian sentiment analysis in e-commerce reviews. Besides all three combined models mentioned above, we consider one more combined model, which is GRU-CNN. We evaluate the performance of each model, then compare the accuracy of the standard models with the combined models to see if the combined models can improve the performance of the standard. Our simulations show that almost all of the hybrid architectures give better accuracies than the standard models. Moreover, the hybrid architecture of LSTM-CNN reaches slightly better accuracies than other hybrid architectures.

Original languageEnglish
Title of host publicationSoft Computing in Data Science - 6th International Conference, SCDS 2021, Proceedings
EditorsAzlinah Mohamed, Bee Wah Yap, Jasni Mohamad Zain, Michael W. Berry
PublisherSpringer Science and Business Media Deutschland GmbH
Pages18-27
Number of pages10
ISBN (Print)9789811673337
DOIs
Publication statusPublished - 2021
Event6th International Conference on Soft Computing in Data Science, SCDS 2021 - Virtual, Online
Duration: 2 Nov 20213 Nov 2021

Publication series

NameCommunications in Computer and Information Science
Volume1489 CCIS
ISSN (Print)1865-0929
ISSN (Electronic)1865-0937

Conference

Conference6th International Conference on Soft Computing in Data Science, SCDS 2021
CityVirtual, Online
Period2/11/213/11/21

Keywords

  • CNN
  • Deep learning
  • GRU
  • LSTM

Fingerprint

Dive into the research topics of 'Performance Analysis of Hybrid Architectures of Deep Learning for Indonesian Sentiment Analysis'. Together they form a unique fingerprint.

Cite this