A Refined Span Classification Model for Recognizing Nested Named Entity in Marine Meteorological Disaster Texts
Open Access
- 10 June 2026
- journal article
- research article
- Published by MDPI AG in ISPRS International Journal of Geo-Information
- Vol. 15 (6) , 258
Abstract
Named entity recognition (NER) in marine meteorological disaster texts is essential for automated information extraction and disaster management. However, disaster-chain descriptions often contain nested entities that are difficult for conventional flat NER models to represent. This paper proposes PRSpan, a position-role-aware span classification model for nested NER. PRSpan incorporates Rotary Position Embedding (RoPE)-enhanced attention for relative position-aware boundary modeling and uses Conditional Layer Normalization (CLN) to generate role-specific Head, Mid, and Tail token features. A Positional Role Pooling strategy further aggregates these features into span representations to preserve boundary cues and internal semantic coherence. To support evaluation, we construct MMD-NER, a domain-specific dataset containing 1899 sentences, 17,017 entities in 11 categories, and 2978 nested entity pairs through a four-step LLM-assisted pipeline. Experimental results show that PRSpan achieves Micro-F1 and Macro-F1 scores of 94.58% and 93.47%, outperforming the strongest baseline by 3.61 and 3.93 percentage points, respectively. Additional analyses verify the effectiveness of RoPE-enhanced attention, role-specific feature generation, and Positional Role Pooling. Cross-domain transfer and LLM prompting comparisons further demonstrate the practical value of PRSpan for nested entity extraction in low-resource Earth science domains.Keywords
Funding Information
- Open Fund Project of the Key Laboratory of Blockchain Agricultural Applications, Ministry of Agriculture and Rural Affairs (2023KLABA01)
- National Science and Technology Major Project (2022ZD0119501)
- Natural Science Foundation of Shandong province, China (ZR2022MF319)
- Natural Science Foundation of Shandong province, China (ZR2025ZD17)
- Taishan Scholar Program of Shandong Province, China (TSTP20250506)
- National Funds through FCT (UID/04152/2025)
- Centro de Investigação em Gestão de Informação (MagIC)/NOVA IMS (UID/PRR/04152/2025)
This publication has 35 references indexed in Scilit:
- A LLM-based agent for the construction of typhoon knowledge graphsEnvironmental Modelling & Software, 2026
- Fine-tuned BERT-BiLSTM-CRF approach for named entity recognition in geological disaster textsEarth Science Informatics, 2025
- Semantic information extraction and search of mineral exploration data using text mining and deep learning methodsOre Geology Reviews, 2024
- RoFormer: Enhanced transformer with Rotary Position EmbeddingNeurocomputing, 2023
- Multi-level context features extraction for named entity recognitionComputer Speech & Language, 2022
- Deep learning-based methods for natural hazard named entity recognitionScientific Reports, 2022
- Chinese Named Entity Recognition in the Geoscience Domain Based on BERTEarth and Space Science, 2022
- Named Entity Recognition by Using XLNet-BiLSTM-CRFNeural Processing Letters, 2021
- Non-Factoid Answer Selection in Indonesian Science Question Answering System using Long Short-Term Memory (LSTM)Procedia Computer Science, 2021
- Automatic spatiotemporal and semantic information extraction from unstructured geoscience reports using text mining techniquesEarth Science Informatics, 2020