MicroConceptBERT: concept-relation based document information extraction framework.

Silva, Kanishka; Silva, Thushari; Nanayakkara, Gayani

Abstract

Extracting information from documents is a crucial task in natural language processing research. Existing information extraction methodologies often focus on specific domains, such as medicine, education or finance, and are limited by language constraints. However, more comprehensive approaches that transcend document types, languages, contexts, and structures would significantly advance the field proposed in recent research. This study addresses this challenge by introducing microConceptBERT: a concept-relations-based framework for document information extraction, which offers flexibility for various document processing tasks while accounting for hierarchical, semantic, and heuristic features. The proposed framework has been applied to a question-answering task on benchmark datasets: SQUAD 2.0 and DOCVQA. Notably, the F1 evaluation metric attains an outperforming 87.01 performance rate on the SQUAD 2.0 dataset compared to baseline models: BERT-base and BERT-large models.

Citation

SILVA, K., SILVA, T. and NANAYAKKARA, G. 2023. MicroConceptBERT: concept-relation based document information extraction framework. In Proceedings of the 7th SLAAI (Sri Lanka Association for Artificial Intelligence) International conference on artificial intelligence 2023 (SLAAI-ICAI 2023), 23-24 November 2023, Kelaniya, Sri Lanka. Piscataway: IEEE [online], article number 10365022. Available from: https://doi.org/10.1109...ICAI59257.2023.10365022

• Presentation Conference Type: Conference Paper (published)
• Conference Name: 7th SLAAI (Sri Lanka Association for Artificial Intelligence) International conference on artificial intelligence 2023 (SLAAI-ICAI 2023)
• Start Date: Nov 23, 2023
• End Date: Nov 24, 2023
• Acceptance Date: Sep 24, 2023
• Online Publication Date: Dec 31, 2023
• Publication Date: Dec 31, 2023
• Deposit Date: Feb 1, 2024
• Publicly Available Date: Feb 1, 2024
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• DOI: https://doi.org/10.1109/SLAAI-ICAI59257.2023.10365022
• Keywords: Concept-relations; Entity extraction; Layout analysis; Ontology; Transformers; Question answering
• Public URL: https://rgu-repository.worktribe.com/output/2225955