BioTagger-GM: A Gene/Protein Name Recognition System

Torii, Manabu; Hu, Zhang-Zhi; Wu, Cathy H.; Liu, Hongfang

doi:10.1197/jamia.m2844

Cited by 55 publications

(50 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…By adding domain knowledge in this form, we could further improve prediction performance (Table 13). The best state-ofthe-art system [30] GM system (to our knowledge) is reported to achieve a slightly higher F1 score of 88.76, however it relies on using very large lexicons (over 15M protein/gene names) with dictionary-based postprocessing and multi-classifier setting. The best performance in Table 13 Table 13: F1 score on the test set for BioCreativeII GM.…”

Section: Comparison With Previous Resultsmentioning

confidence: 99%

Semi-Supervised Bio-Named Entity Recognition with Word-Codebook Learning

Kuksa¹,

Qi²

2010

Proceedings of the 2010 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

We describe a novel semi-supervised method called WordCodebook Learning (WCL), and apply it to the task of bionamed entity recognition (bioNER). Typical bioNER systems can be seen as tasks of assigning labels to words in bioliterature text. To improve supervised tagging, WCL learns a class of word-level feature embeddings to capture word semantic meanings or word label patterns from a large unlabeled corpus. Words are then clustered according to their embedding vectors through a vector quantization step, where each word is assigned into one of the codewords in a codebook. Finally codewords are treated as new word attributes and are added for entity labeling. Two types of wordcodebook learning are proposed: (1) General WCL, where an unsupervised method uses contextual semantic similarity of words to learn accurate word representations; (2) Task-oriented WCL, where for every word a semi-supervised method learns target-class label patterns from unlabeled data using supervised signals from trained bioNER model. Without the need for complex linguistic features, we demonstrate utility of WCL on the BioCreativeII gene name recognition competition data, where WCL yields state-of-the-art performance and shows great improvements over supervised baselines and semi-supervised counter peers.

show abstract

Section: Comparison With Previous Resultsmentioning

confidence: 99%

Semi-Supervised Bio-Named Entity Recognition with Word-Codebook Learning

Kuksa¹,

Qi²

2010

Proceedings of the 2010 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

show abstract

“…J-D Kim et al [12] used supervised learning approaches which require the annotated corpora for the development , evaluation of Relation Extraction and shown good performance. Manabu Torii et al [18] developed BioTagger-TM using i)rule/pattern based recognition methods characterized by handcrafted name/context patterns and associated rules ii) dictionary look up methods requiring a list of entity names and iii)Machine learning methods utilizing named entity tagged corpora . In their work on large entity corpus, machine learning methods had given promising performance.…”

Section: Ner Using Hybrid Methodsmentioning

confidence: 99%

Biomedical Named Entity Recognition - a swift review

Vijaya¹

2017

ijecs

View full text Add to dashboard Cite

Abstract:The main focus of this paper is taking a swift review on the Biomedical Named Entity Recognition which is the most complex task in Information Extraction. This paper analyses various methods used for NER particularly in the field of Biomedical domain. The aim of this study is to discuss about the methods used to recognize Biological entities like genes, proteins and diseases etc., and propose an effective method to recognize heterogeneous entities.

show abstract

“…31 It was developed further after the workshop and tested in our evaluation study. 13 It has been used for biological literature mining. 37 38 BioTagger-GM is essentially a machine learning tagger exploiting features based on dictionary lookup.…”

Section: Biotagger-gmmentioning

confidence: 99%

“…The performance of BioTagger-GM is owing to BioThesaurus and the UMLS Metathesaurus, as well as the powerful machine learning algorithms as reported in our previous study. 13 In the current task, instead of using BioThesaurus, we used a collection of clinical terms extracted from discharge summaries for a clinical vocabulary viewer. 45 The UMLS Metathesaurus was used just as in BioTagger-GM.…”

Section: Adaptation Of Biotagger-gmmentioning

confidence: 99%

Using machine learning for concept extraction on clinical documents from multiple data sources

Torii¹,

Wagholikar²,

Liu³

2011

J Am Med Inform Assoc

106

View full text Add to dashboard Cite

Objective Concept extraction is a process to identify phrases referring to concepts of interests in unstructured text. It is a critical component in automated text processing. We investigate the performance of machine learning taggers for clinical concept extraction, particularly the portability of taggers across documents from multiple data sources. Methods We used BioTagger-GM to train machine learning taggers, which we originally developed for the detection of gene/protein names in the biology domain. Trained taggers were evaluated using the annotated clinical documents made available in the 2010 i2b2/VA Challenge workshop, consisting of documents from four data sources. Results As expected, performance of a tagger trained on one data source degraded when evaluated on another source, but the degradation of the performance varied depending on data sources. A tagger trained on multiple data sources was robust, and it achieved an F score as high as 0.890 on one data source. The results also suggest that performance of machine learning taggers is likely to improve if more annotated documents are available for training. Conclusion Our study shows how the performance of machine learning taggers is degraded when they are ported across clinical documents from different sources. The portability of taggers can be enhanced by training on datasets from multiple sources. The study also shows that BioTagger-GM can be easily extended to detect clinical concept mentions with good performance.

show abstract

BioTagger-GM: A Gene/Protein Name Recognition System

Abstract: The results suggest that terminology sources, powerful machine learning frameworks, and system combination can be integrated to build an effective BNER system.

Cited by 55 publications

References 34 publications

Semi-Supervised Bio-Named Entity Recognition with Word-Codebook Learning

Semi-Supervised Bio-Named Entity Recognition with Word-Codebook Learning

Biomedical Named Entity Recognition - a swift review

Using machine learning for concept extraction on clinical documents from multiple data sources

Contact Info

Product

Resources

About