Current Methodologies for Biomedical Named Entity Recognition
Current Methodologies for Biomedical Named Entity Recognition
复制标题
当前生物医学命名实体识别方法
DOI：
10.1002/9781118617151.ch37
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
J. Oliveira
中科院分区：
文献类型：
--
作者：
David Campos;Sérgio Matos;J. Oliveira
文本挖掘的主要目标是检索隐藏在文本中的知识，并以简明、简单的形式将其呈现给最终用户。为了实现这一目标，可以定义两个主要的研究方向：信息提取(IE)和信息检索(IR)。命名实体识别(NER)是最重要的任务之一，因为IE步骤将使用它提供的名称执行。本章分别介绍了使用词典、机器学习和混合方法实现解决方案所需的步骤。文中给出了每种方法的一些实例。它还描述了开发NER和标准化解决方案的现有方法，并结合一些现有系统的实例介绍和解释了核心技术。本章为每种方法提供一个实际例子，以说明如何执行这几个步骤。
The primary goal of text mining is to retrieve knowledge that is hidden in text and to present it in a concise and simple form to the final users. To achieve this objective, two main directions of research can be defined: Information Extraction (IE) and Information Retrieval (IR). Named Entity Recognition (NER) is one of the most important tasks, since the IE steps will be performed using the names provided by it. The chapter presents the steps necessary to implement solutions using dictionaries and machine learning and hybrid approaches, respectively. Some practical examples for each approach are provided. It also describes the existing approaches to develop NER and normalization solutions, presenting and explaining the core techniques accompanied with examples of some existent systems. The chapter presents one practical example for each approach in order to demonstrate how the several steps could be implemented.