半结构化数据的结构发现聚类算法

当处理的半结构数据时,从中挖掘出其隐含的结构显得愈来愈重要了,不仅对于用户而且对于整个系统来说,其重要性也是不言而喻的.算法structure_discovery将聚类算法应用到半结构化数据的结构发现上,将描述同一类对象的共有结构按照相似度聚集到一起形成类,然后从类中发现满足用户兴趣度的结构.与同类算法比较,structure_discovery为增量式算法,同时,聚类技术的使用提高了结构发现效率.

Clustering Algorithm on Structure Discovery of Semi-structured Data

When processing semi-structured data,finding out its underlying structures is increasingly important,not only for users but also for the entire system.Structure_discovery applied the clustering algorithm to the structure discovery of semi-structured data,aggregated the common structure of the same type of objects according to similarity to form a class,and then found the structure meeting the user interest from the class.Compared with similar algorithms,structure_discovery is the incremental algorithm,and t...