AI-woordenlijst
Het complete woordenboek van kunstmatige intelligentie
NER by Rules and Patterns
Traditional approach using linguistic rules and regex patterns to identify named entities.
NER Statistical
Method based on statistical models such as CRF (Conditional Random Fields) for entity recognition.
NER by Deep Learning
Use of deep neural networks (LSTM, BiLSTM-CRF, Transformers) for automatic entity recognition.
Multilingual NER
Systems capable of recognizing named entities in multiple languages with a unified model or specialized models.
Domain-Specific NER
Adapting NER for specific domains such as medical, legal, financial, or scientific with particular entities.
NER for Nested Entities
Advanced technique for detecting named entities that overlap or are contained within each other.
NER by Transfer Learning
Approach using pre-trained models (BERT, RoBERTa) fine-tuned for specific named entity recognition tasks.
NER with Weak Supervision
Semi-supervised or weakly-supervised learning methods requiring few annotated data to train NER models.
NER for Unstructured Data
Specialization of NER to handle complex documents such as emails, reports, or PDF documents with varied structures.
Real-time NER
Systems optimized for entity recognition in continuous data streams with strict latency constraints.
NER for Social Networks
Adaptation of NER to handle informal language, emojis, and specific features of social media texts.
NER by Ontology Alignment
An approach integrating external knowledge through ontologies and knowledge bases to improve entity recognition.