Textual Entailment Technique for the Bahasa Using BiLSTM

I. Made Suwija Putra, Daniel Siahaan*, Ahmad Saikhu

*Corresponding author for this work

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

3 Citations (Scopus)

Abstract

Recognizing Textual Entailment (RTE) is an important task in Natural Language Understanding (NLU). RTE aims to identify the entailment relationship between two text fragments (premise and hypothesis). Researchers on textual entailment have converged various languages. Research indicates that the use of deep learning approaches improved the accuracy performance of RTE in English. This study proposes an RTE model in Bahasa using a Bidirectional Long Short-Term Memory (BiLSTM). The method was trained on a large corpus in Bahasa. The corpus was generated by translating the SNLI corpus into Bahasa (SNLI Indo). The experimentation shows that the best training accuracy that the model produced, i.e., 79.37%, is when BiLSTM was combined with pre-trained IndoBERT as the word embedding method.

Original languageEnglish
Title of host publication2022 International Seminar on Intelligent Technology and Its Applications
Subtitle of host publicationAdvanced Innovations of Electrical Systems for Humanity, ISITIA 2022 - Proceeding
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages139-144
Number of pages6
ISBN (Electronic)9781665460811
DOIs
Publication statusPublished - 2022
Event23rd International Seminar on Intelligent Technology and Its Applications, ISITIA 2022 - Virtual, Surabaya, Indonesia
Duration: 20 Jul 202221 Jul 2022

Publication series

Name2022 International Seminar on Intelligent Technology and Its Applications: Advanced Innovations of Electrical Systems for Humanity, ISITIA 2022 - Proceeding

Conference

Conference23rd International Seminar on Intelligent Technology and Its Applications, ISITIA 2022
Country/TerritoryIndonesia
CityVirtual, Surabaya
Period20/07/2221/07/22

Keywords

  • Bahasa
  • BiLSTM
  • Recognizing Textual Entailment
  • SNLI Indo
  • Word Embedding

Fingerprint

Dive into the research topics of 'Textual Entailment Technique for the Bahasa Using BiLSTM'. Together they form a unique fingerprint.

Cite this