Reinforcement learning based distantly supervised biomedical named entity recognition

Data used by current Biomedical named entity recognition (BioNER) systems has mostly been manually labelled for supervision. However, it might be difficult to find large amounts of annotated data, especially in fields with a high level of specialization, such as biomedical, bioinformatics, and so on. When dictionaries and ontologies are available, which are domain-specific knowledge resources, automatically tagged distantly supervised biomedical training data can be developed. However, any such distantly supervised NER result is normally noisy. The prevalence of false positives and false negatives with this type of autonomously generated data is the main problem that directly affects efficiency. This research investigates distant supervision to detect false positive occurrences in BioNER task. A reinforcement learning technique is employed that is modelled as a graphical processing unit (GPU) accelerated Markov decision process (MDP) with a neural network policy. To deal with false negative cases, we employ a partial annotation conditional random field (CRF) technique. Results on two benchmark datasets show a cutting-edge methodology that can enhance the functionality of the neural NER system. It goes on to show how the proposed approach cuts down on human annotated data for BioNER tasks in Natural Language Processing (NLP).

• Publication year

  2022

• Venue

  International Journal of Intelligent Decision Technologies

• Publication date

  2022-03-07

• Fields of study

  Medicine, Computer Science

• Identifiers
  DOI 10.3233/IDT-220205
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Reinforcement-based denoising of distantly supervised NER with partial annotation

  2019cited by this paper
• SciBERT: A Pretrained Language Model for Scientific Text

  2019cited by this paper
• Distantly Supervised Biomedical Named Entity Recognition with Dictionary Expansion

  2019cited by this paper
• Cross-type Biomedical Named Entity Recognition with Deep Multi-Task Learning

  2018cited by this paper
• Distantly Supervised NER with Partial Annotation Learning and Reinforcement Learning

  2018cited by this paper
• Contextual String Embeddings for Sequence Labeling

  2018cited by this paper
• Deep Contextualized Word Representations

  2018cited by this paper
• Robust Distant Supervision Relation Extraction via Deep Reinforcement Learning

  2018cited by this paper
• Reinforcement Learning for Relation Classification From Noisy Data

  2018cited by this paper
• Learning Named Entity Tagger using Domain-Specific Dictionary

  2018cited by this paper
• Empower Sequence Labeling with Task-Aware Neural Language Model

  2017cited by this paper
• Automated Phrase Mining from Massive Text Corpora

  2017cited by this paper
• SwellShark: A Generative Model for Biomedical Named Entity Recognition without Labeled Data

  2017cited by this paper
• End-to-end Sequence Labeling via Bi-directional LSTM-CNNs-CRF

  2016cited by this paper
• Neural Architectures for Named Entity Recognition

  2016influential reference
• Global Distant Supervision for Relation Extraction

  2016cited by this paper
• ClusType: Effective Entity Recognition and Typing by Relation Phrase-Based Clustering

  2015cited by this paper
• Relation Extraction from the Web Using Distant Supervision

  2014cited by this paper
• Relation Extraction with Matrix Factorization and Universal Schemas

  2013cited by this paper
• Markov Decision Process Parallel Value Iteration Algorithm On GPU

  2013cited by this paper
• Multi-instance Multi-label Learning for Relation Extraction

  2012cited by this paper
• Pattern Learning for Relation Extraction with a Hierarchical Topic Model

  2012cited by this paper
• Towards mature use of semantic resources for biomedical analyses

  2011cited by this paper
• Modeling Relations and Their Mentions without Labeled Text

  2010cited by this paper
• Design Challenges and Misconceptions in Named Entity Recognition

  2009cited by this paper
• An Intelligent Assistant for Power Plants Based on Factored MDPs

  2009cited by this paper
• GPU-based Markov decision process solver

  2009cited by this paper
• Distant supervision for relation extraction without labeled data

  2009cited by this paper
• Training Conditional Random Fields Using Incomplete Annotations

  2008cited by this paper
• A survey of named entity recognition and classification

  2007cited by this paper
• Parallel Markov Decision Processes

  2007cited by this paper
• Simple Statistical Gradient-Following Algorithms for Connectionist Reinforcement Learning

  2004cited by this paper
• Predictive autonomous robot navigation

  2002cited by this paper
• Policy Gradient Methods for Reinforcement Learning with Function Approximation

  1999cited by this paper
• Solving Very Large Weakly Coupled Markov Decision Processes

  1998cited by this paper
• Prioritized Goal Decomposition of Markov Decision Processes: Toward a Synthesis of Classical and Decision Theoretic Planning

  1997cited by this paper
• How to Dynamically Merge Markov Decision Processes

  1997cited by this paper
• Acting under uncertainty: discrete Bayesian models for mobile-robot navigation

  1996cited by this paper

• No citing papers are available for this paper.