微博文本和传统文本体裁特征对比

Comparison of Genre Features Between Micro-blog Text and Traditional Text

在总结常用特征集合的基础上,根据微博文本的特点以及特征选取原则,选取了适合微博文本体裁分析的特征集合,这些特征能典型的反应微博文本和其他文本形式的区别.还分别对不同的文本体裁进行特征值的统计,并将统计结果在不同的文本体裁之间进行了深入的对比分析,并从体裁的角度分析出不同文本体裁的特征值差别的原因.并从体裁特征的角度说明微博文本是一种新的体裁文本.

After summarizing common feature set, according to micro-blog text＇s characteristics and feature selection principles, this paper selects some feature sets which are suitable for micro-blog text genre analysis. These features can typically reflect the differences between micro-blog text and other text forms. It performs a statistical analysis on different text genre ,and inputs the statistical result to comparison analysis among various text genres, and finds out reasons for characteristic value differences from the perspective of genre feature. It also proves that micro-blog text is a new text genre.