We first segment a news page into various blocks using VIPS algorithm and extract visual features and content features of the block. The title block dwells on a manifold in the resulted feature space. 在使用VIPS算法对新闻网页分块的基础上,我们抽取新闻标题块的视觉特征和部分内容特征,构造了一个标题块数据的流形空间。