The quick and intelligent requests and answers in artificial intelligence (AI) are inseparable from intelligent data. Knowledge graph makes data more intelligent by establishing association among data, which provides convenience for intelligent search, reasoning and analysis of data. Resource Description Framework (RDF) is an effective data representation model of knowledge graph. This paper takes RDF as the research object and proposes an incremental partition method of intelligent data (IPID) to realize the distributed storage of large-scale AI data. First, IPID gives a mixed object function integrating edge cut and load balancing. Second, IPID devises the initial and incremental partitioning algorithms of RDF. The initial partition divides the original RDF graph into kernel vertices, boundary vertices and free vertices. The boundary and freedom nodes select the kernel vertex with the maximum gain of object function to form a sub-partition. And the incremental partition is in charge of the selection of sub-partition of new and deleted data by the object function. Meanwhile, the incremental partition algorithm would also execute a dynamic adjustment strategy at a certain time interval according to the balance and tightness of sub-partition to satisfy the partitioning object. Finally, IPID is tested on the knowledge graph datasets. The experimental results show that the object function guarantees the quality of knowledge graph partition in edge cut and load balancing, and effectively realizes the incremental partition.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.