Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
September 10, 2025Journal of Information Systems Engineering and Business IntelligenceOpen Access

Incorporation of IndoBERT and Machine Learning Features to Improve the Performance of Indonesian Textual Entailment Recognition

View Full Paper
Ask AI
Bookmark
Share

Authors

TTTeuku Yusransyah TandiTATaufik Fuadi AbidinHRHammam Riza

Discussion

Loading...

Member takes

Overview

This hybrid model enhances F1-score and computational efficiency in Indonesian NLP tasks, suggesting significant improvements in data processing.

Key Points

  • Hybrid-IndoBERT-RTE achieved an F1-score of 85%, indicating a strong performance in textual entailment tasks.
  • The model required 4.2 times less GPU VRAM, highlighting significant reductions in computational resource demands.
  • Training was up to 44.44 times more efficient than IndoBERT-large-p1, indicating a substantial improvement in efficiency.
  • Future work will focus on expanding and diversifying the datasets used for Indonesian textual entailment recognition.

Cite This Study

Tandi et al. (2025) studied this question.

synapsesocial.com/papers/68c1ac0154b1d3bfb60e482fhttps://doi.org/10.20473/jisebi.11.2.173-186
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1EVALUATION OF INDOBERT AND ROBERTA: PERFORMANCE OF INDONESIAN LANGUAGE TRANSFORMER MODELS IN SENTIMENT CLASSIFICATION2025 · 2 citations
  2. 2COMPARATIVE PERFORMANCE OF TRANSFORMER AND LSTM MODELS FOR INDONESIAN INFORMATION RETRIEVAL WITH INDOBERT2025
  3. 3Optimizing IndoBERT for Revised Bloom's Taxonomy Question Classification Using Neural Network Classifier2025 · 2 citations
  4. 4Bridging the Lingustic Gap: Challenges in Building AI Models For Non-Standard Dialects2025 · 1 citations
  5. 5Implementation of Text Mining for Evaluating the Relevance Between News Headlines and Content on a Web-Based Platform2025