Skip to main content

Table 1 Performance in the GM task

From: Automating curation using a natural language processing pipeline

Run

Method

Heldout

Test

  

Precision

Recall

F1

Precision

Recall

F1

1

CRF

0.8594

0.8211

0.8398

0.8697

0.8255

0.8470

2

BMEMM

0.8597

0.7982

0.8278

0.8638

0.8041

0.8329

3

CRF

0.8463

0.8297

0.8379

0.8649

0.8248

0.8444

  1. BMEMM, bidirectional maximum entropy Markov model; CRF, conditional random field; GM, gene mention.
Back to article page

Genome Biology

ISSN: 1474-760X

Contact us