摘要
文章意义段的自动划分技术是自然语言理解研究领域中的一个非常重要的研究课题。该文在对文章意义段划分进行研究与实践的基础上,提出了由计算机自动划分意义段的数学模型。通过计算文本中用词重复数,建立用词重复频率三角矩阵,给出了各个自然段归并成意义段的制约条件。实践证明,该数学模型反映了一类文章的客观结构。
The technology of the automatic parting text meaning paragraph is an extremely important research task in natural language understanding field. This papar divides the text into the meaning paragraph on the basis of research and practice.This papar proposes mathematical model of automatic parting text meaning paragraph with computer, calculates word frequency of the text with computer, builds triangular matrix of reused word frequency, gives restricted condition of generated meaning paragraph. By practice, it＇s proved that this mathematical model presents objective structure of some text.
出处
《计算机工程》
CAS
CSCD
北大核心
2007年第13期205-206,共2页
Computer Engineering
关键词
意义段
词频
三角矩阵
meaning paragraph
word-frequency
triangular matrix
