NikosKprl
/

en_deberta_v3_base_ner_method

Token Classification

Eval Results (legacy)

Model card Files Files and versions

A Named Entity Recognition (NER) model specialized in detecting methods or techniques from scholarly publications.

Feature	Description
Name	`en_deberta_v3_base_ner_method`
Version	`0.1.0`
spaCy	`>=3.8.7,<3.9.0`
Default Pipeline	`transformer`, `ner`
Components	`transformer`, `ner`
License	MIT

Label Scheme

View label scheme (1 labels for 1 components)

Component	Labels
`ner`	`METHOD`

Accuracy

Type	Score
`ENTS_F`	81.08
`ENTS_P`	81.40
`ENTS_R`	80.75

Downloads last month: 15

Model tree for NikosKprl/en_deberta_v3_base_ner_method

Base model

microsoft/deberta-v3-base

Finetuned

(612)

this model

Collection including NikosKprl/en_deberta_v3_base_ner_method

Research Spotlight

Research Spotlight is a modular workflow written in Python for converting research publications into an ontology-driven scholarly knowledge graph. • 3 items • Updated Apr 8

Evaluation results

NER Precision
self-reported

0.814
NER Recall
self-reported

0.808
NER F Score
self-reported

0.811