摘要
是一种基于计算机的系统,它概括文本的同时保留文章的主题。在获取摘要过程中,用句子间的权重给每个段落建立句子的图谱;同时也考虑印尼文章段落结构的归纳演绎方法,用最短路径算法确定哪些句子部分将成为摘要的结果。实验结果表明,Dijkstra算法优于Bellman-Ford算法生成文本摘要压缩率的12%。
Automated Text Summarization is a computer-based system to perform text summarization that still keeps the main subject of the article. In the summarization process, we use weight of sentence, and build a graph of sentences in every paragraph. We also consider the inductive-deductive method in paragraph structure in Indonesian articles. We use shortest part algorithm to determine which sentences will become results of summarization. In this paper, we will compare the results of summarization with Dijkstra and bellman-ford algorithms. The experimental result shows that Dijkstra give the better compression rate by 12%.
出处
《科技广场》
2015年第7期16-20,共5页
Science Mosaic
