Deep Learning for Detecting Entailment Between Requirements Using Semantics from Use Case Diagrams as Training Data: A Comparative Study

Dony Bahtera Firmawan, Daniel Siahaan*, Ahmad Saikhu

*Corresponding author for this work

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

Abstract

Textual entailment, also known as natural language inference, is a branch of natural language processing (NLP) that examines the semantics and meaning of phrases and text excerpts in order to ascertain whether a hypothesis may be drawn from a premises. This study examines how well five distinct deep learning techniques work (BiLSTM, RoBERTa, GRU, Bi-GRU, and MLP) using different word embeddings (GloVe and BERT) in identifying textual entailment in SRS documents. The dataset used in this research was obtained from the use case diagrams (UCD) and the use case specifications (UCS) written in English and the SNLI corpus. Overall, the BiLSTM-GloVe model achieved the best results compared to other models on the proposed dataset with an accuracy of 49%, an F1-score of 46%, a precision of 46%, and a recall of 59%.

Original languageEnglish
Title of host publication2024 International Seminar on Intelligent Technology and Its Applications
Subtitle of host publicationCollaborative Innovation: A Bridging from Academia to Industry towards Sustainable Strategic Partnership, ISITIA 2024 - Proceeding
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages669-674
Number of pages6
Edition2024
ISBN (Electronic)9798350378573
DOIs
Publication statusPublished - 2024
Event25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024 - Hybrid, Mataram, Indonesia
Duration: 10 Jul 202412 Jul 2024

Conference

Conference25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024
Country/TerritoryIndonesia
CityHybrid, Mataram
Period10/07/2412/07/24

Keywords

  • Comparative Study
  • Deep Learning
  • SRS Documents
  • Text Entailment
  • Use Case Diagram

Fingerprint

Dive into the research topics of 'Deep Learning for Detecting Entailment Between Requirements Using Semantics from Use Case Diagrams as Training Data: A Comparative Study'. Together they form a unique fingerprint.

Cite this