【摘要】 本文简要介绍了文本的向量空间模型表示,并提出了用该模型进行非受限真实文本的自动文摘方法。其基本思想是对输入文本进行项(本文为词和概念)的统计,根据不同项在文本中的分布评估其重要性,据此和其它信息形成句子的综合权重,生成摘要。文中较详细地介绍了系统的实现及实验结果,同时提出了对文摘系统的评价方法。更多还原

【Abstract】 Based on the vector space model (VSM),we suggest an approach to automatic abstraction from real unconstrained text. Since all terms in one text are not equally useful for content representation, a term-weighting system is introduced that assigns higher weights to more important terms and lower weights to less important terms. This paper describes the implementation and detailed experimental results of the system,and some methods to evaluate the results.更多还原