Exploiting context for biomedical entity recognition

Finkel, Jenny Rose; Dingare, Shipra; Nguyen, Huy Van; Nissim, Malvina; Manning, Christopher D.; Sinclair, Gail

doi:10.3115/1567594.1567614

Cited by 102 publications

(86 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We selected one of the feature sets, F2, as an orthographic feature that was utilized as the best indicator in biomedical NER. There are many studies based on orthographic features [8][9][10][11][12][13][14][15][16][17][18][19][20] and we found the best to be [11]. Another set of features, F1, was extracted as a context-based feature set with lexicons, which provides domain knowledge in a consistent manner.…”

Section: Data Preprocessing and Feature Processing For Biomedical Nermentioning

confidence: 99%

See 1 more Smart Citation

An Active Co-Training Algorithm for Biomedical Named-Entity Recognition

Munkhdalai¹,

Li²,

Yun³

et al. 2012

Journal of Information Processing Systems

View full text Add to dashboard Cite

Abstract-Exploiting unlabeled text data with a relatively small labeled corpus has been an active and challenging research topic in text mining, due to the recent growth of the amount of biomedical literature. Biomedical named-entity recognition is an essential prerequisite task before effective text mining of biomedical literature can begin. This paper proposes an Active Co-Training (ACT) algorithm for biomedical named-entity recognition. ACT is a semi-supervised learning method in which two classifiers based on two different feature sets iteratively learn from informative examples that have been queried from the unlabeled data. We design a new classification problem to measure the informativeness of an example in unlabeled data. In this classification problem, the examples are classified based on a joint view of a feature set to be informative/non-informative to both classifiers. To form the training data for the classification problem, we adopt a query-bycommittee method. Therefore, in the ACT, both classifiers are considered to be one committee, which is used on the labeled data to give the informativeness label to each example. The ACT method outperforms the traditional co-training algorithm in terms of fmeasure as well as the number of training iterations performed to build a good classification model. The proposed method tends to efficiently exploit a large amount of unlabeled data by selecting a small number of examples having not only useful information but also a comprehensive pattern.

show abstract

Section: Data Preprocessing and Feature Processing For Biomedical Nermentioning

confidence: 99%

“…Since the machine learning approach was adopted, significant progress in biomedical NER has been achieved with methods like the Markov Model [7], the Support Vector Machine (SVM) [8][9][10][11][12][13] the Maximum Entropy Markov Model [14], and Conditional Random Fields (CRF) [15][16][17][18][19].…”

Section: Related Workmentioning

confidence: 99%

An Active Co-Training Algorithm for Biomedical Named-Entity Recognition

Munkhdalai¹,

Li²,

Yun³

et al. 2012

Journal of Information Processing Systems

View full text Add to dashboard Cite

show abstract

“…Head nouns were used in [37] and [38]; word lists that are highly associated to classes are extracted as lexicons in [52]; keyword lexicons are statistically computed in [39]; keyword and boundary lists in [50].…”

Section: Jnlpba'04 Corpus and Current Solutionsmentioning

confidence: 99%

Towards a Protein–Protein Interaction information extraction system: Recognizing named entities

Dánger

Pla

Molina

et al. 2014

Knowledge-Based Systems

View full text Add to dashboard Cite

The majority of biological functions of any living being are related to ProteinProtein Interactions (PPI). PPI discoveries are reported in form of research publications whose volume grows day after day. Consequently, automatic PPI information extraction systems are a pressing need for biologists. In this paper we are mainly concerned with the named entity detection module of PPIES (the PPI Information extraction system we are implementing) which recognizes twelve entity types relevant in PPI context. It is composed of two sub-modules: a dictionary look-up with extensive normalization and acronym detection, and a Conditional Random Field classifier. The dictionary look-up module has been tested with Interaction Method Task (IMT), and it improves by approximately 10% the current solutions that do not use Machine Learning (ML). The second module has been used to create a classifier using the Joint Workshop on Natural Language Processing in Biomedicine and its Applications (JNLPBA'04) data set. It does not use any external resources, or complex or ad-hoc post-processing, and obtains 77.25%, 75.04% and 76.13 for precision, recall, and F1-measure, respectively, improving all previous results obtained for this data set.

show abstract

“…In 2004, the JNLPBA [20] open challenge task for bio-NER simplified the 36 entity classes in the GENIA corpus [21] and used only five classes, namely protein, DNA, RNA, cell line, and cell type, to evaluate the performance of the participating systems. Unlike the earliest rule-based NER system [14], the following four types of classification models were applied by the participating teams: Support Vector Machines (SVMs) [16,34], Hidden Markov Models (HMMs) [50], Maximum Entropy Markov Models (MEMMs) [13] and Conditional Random Fields (CRFs) [38]. The most frequently applied models were SVMs.…”

Section: Biological Named Entity Recognitionmentioning

confidence: 99%