Sentiment Analysis Using State of the Art Machine Learning Techniques

Creative Commons License

Balci S., Demirci G. M., Demirhan H., Sarp S.

9th Machine Intelligence and Digital Interaction Conference, MIDI 2021, Virtual, Online, 9 - 10 December 2021, vol.440 LNNS, pp.34-42, (Full Text)

  • Publication Type: Conference Paper / Full Text
  • Volume: 440 LNNS
  • Doi Number: 10.1007/978-3-031-11432-8_3
  • City: Virtual, Online
  • Page Numbers: pp.34-42
  • Keywords: Bag of tricks, BERT, CNN, Sentiment analysis, Transformer
  • Open Archive Collection: AVESIS Open Access Collection
  • Yozgat Bozok University Affiliated: No

Abstract

Sentiment analysis is one of the essential and challenging tasks in the Artificial Intelligence field due to the complexity of the languages. Models that use rule-based and machine learning-based techniques have become popular. However, existing models have been under-performing in classifying irony, sarcasm, and subjectivity in the text. In this paper, we aim to deploy and evaluate the performances of the State-of-the-Art machine learning sentiment analysis techniques on a public IMDB dataset. The dataset includes many samples of irony and sarcasm. Long-short term memory (LSTM), bag of tricks (BoT), convolutional neural networks (CNN), and transformer-based models are developed and evaluated. In addition, we have examined the effect of hyper-parameters on the accuracy of the models.