潜在语义分析聚类算法在文摘句子排序中的应用

针对多文档文摘生成过程中话题容易中断和文摘句子语义出现不连贯这两个研究难点, 分析了潜在语义分析聚类算法在句子排序中的应用, 以期提高文摘的生成质量。先采用潜在语义分析聚类算法将文摘句子聚类, 从而形成话题集, 以达到解决话题中断的目的。通过计算文档的文摘展现力, 挑选出文摘展现力最大的文档作为模板, 然后根据模板对文摘句子进行两趟排序。实验结果表明, 提出的算法是有效的, 该算法能够提高文摘的可读性。

Application of latent semantic analysis clustering algorithm in summary sentence ordering

In order to improve the performance of summarization, this paper researched on application of latent semantic analysis clustering algorithm in summary sentence ordering. It adopted latent semantic analysis clustering algorithm to cluster summary sentences into topic sets to make the sentences within the same topic without be interrupted. It selected the ordering template by the summary representation of document, and ordered topic sets by the relational position in the template, ordered the sentences in the same topic set by the position in template. The experimental result shows that the proposed method is effective. It improves the readability of the summary.