基于图形数据库的DBLP数据存储

摘要

DBLP data has large data scale and includes complex relationships including co-authorship relationships,writing relationships, citation relationships and so on. Relational database will lead to the data redundancy and is difficult to update. Propose a DBLP method of data presentation and storage based on graphics database to solve the problem of efficient storage and dynamic updating of complex rela-tionship. Papers,authors,books and other entities are stored by node,and the various relationships between the entities are stored by edge. The experimental results show that this method can effectively support the dynamic additions and deletions,multi-step queries,depth trav-ersal,and breadth traversal operation. It is an effective solution for the complex relationship data storage.%DBLP数据不但数据规模大，而且数据之间存在多种类型的关联关系，包括合作关系、写作关系、引用关系等。采用关系数据存储时不但存在大量的数据冗余，并且难于动态更新，为解决复杂关联关系的高效存储与动态更新问题，提出一种基于图形数据库的DBLP数据表示与存储方法，论文、作者、书籍等实体以节点存储，而实体间的各种关系以多类型的边存储。实验表明该方法能够有效支持关联关系的动态增删、多阶查询、深度遍历、广度遍历等操作，有效解决了复杂关联关系的数据存储问题。