中文摘要：

传统的英文文本语块识别的方法大多是通过设定相应的短语标识符号，最终把语块识别问题转化成词性标注问题来解决．实验表明，这种方法不能充分考虑相邻词性的关系和每种短语的内部组成规律．关联度评价中心词扩展的英文文本语块识别方法从另外一个角度来识别英文文本语块．它具有以下特点：①把每个短语看成是以中心词为核心的聚簇，充分考虑每种短语的内部组成规律；②使用关联度和可信度动态地评价得到的结果．通过对公共测试集的测试，此方法识别的速度较快，而且英语语块识别的F测度值达到了94．05％，与目前的最好结果相当．

英文摘要：

Traditional English text chunking approach is to transfer chunking to part of speech. It is shown that this could not take into account the relationship of neighbor part of speech and the cohesion of all part of speeches within one phrase. In this paper, the headword extending and the evaluation of relative-degree strategy are proposed and applied in the identification of English text chunking whose main features are： 1） regarding each phrase as a cluster whose kernel is headword, which richly uses the disciplinarian of consisting of one phrase; 2） dynamically evaluating the chunking result using doubt-degree and reliability. Through testing on the public corpus, the speed of this method is faster than others, and the F score achieves 94.05 %, which is at the state-of-the-art.