生物医学文献中的疾病命名实体识别问题是疾病相关的生物信息学分析基础,疾病命名实体中的医学术语识别和边界确定是该问题的难点和关键。文中提出了一种CRF(Conditional Random Field)与词典相结合的疾病命名实体识别方法。该方法利用网络资源来构建含有语义信息的医学术语词典,并使用该词典对医学术语进行识别,获得医学术语的语义信息,然后CRF结合这些信息对疾病命名实体进行识别。实验结果表明该方法有效。
In general,finding a feasible route subject to multiple additive QoS constraints is an NP-complete problem.In this paper,we first analyze the constraints and their properties that can reflect the basic characteristics of a networkand support the fundamental QoS requirements. Then,we show that the unicast QoS routing problem can be general-ized as a Multiple Constrained Path(MCP) or Restricted Shortest Path(RSP) problem. Finally,We investigate the ef-ficient approximate algorithms for MCP and RSP,and how to convert a RSP problem to an easier MCP problem.
