生物文本探索：一个基于网络的生物医学文本挖掘套件，用于概念发现。

BioTextQuest: a web-based biomedical text mining suite for concept discovery.

机构信息

Bioinformatics Research Laboratory, Department of Biological Sciences, University of Cyprus, CY 1678, Nicosia, Cyprus, Greece.

出版信息

Bioinformatics. 2011 Dec 1;27(23):3327-8. doi: 10.1093/bioinformatics/btr564. Epub 2011 Oct 12.

DOI:10.1093/bioinformatics/btr564

PMID:21994227

Abstract

SUMMARY

BioTextQuest combines automated discovery of significant terms in article clusters with structured knowledge annotation, via Named Entity Recognition services, offering interactive user-friendly visualization. A tag-cloud-based illustration of terms labeling each document cluster are semantically annotated according to the biological entity, and a list of document titles enable users to simultaneously compare terms and documents of each cluster, facilitating concept association and hypothesis generation. BioTextQuest allows customization of analysis parameters, e.g. clustering/stemming algorithms, exclusion of documents/significant terms, to better match the biological question addressed.

AVAILABILITY

http://biotextquest.biol.ucy.ac.cy

CONTACT

vprobon@ucy.ac.cy; iliopj@med.uoc.gr

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

摘要

BioTextQuest 通过命名实体识别服务，将文章聚类中重要术语的自动发现与结构化知识注释相结合，提供了交互友好的可视化界面。基于标签云的术语展示对每个文档聚类进行语义标注，根据生物实体进行标注，并且提供文档标题列表，使用户能够同时比较每个聚类的术语和文档，促进概念关联和假设生成。BioTextQuest 允许自定义分析参数，例如聚类/词干算法、排除文档/重要术语，以更好地匹配所解决的生物学问题。