Research And Application Of Social Network Data Based On Neo4j Graph Database

As one of the new,practical dating mode,social network has attracted majority of users of all ages,and is playing an increasingly important role in the network activity.Since social network is frequently used,and involves many people,it generates massive amounts of data.Traditional relational database cannot handle massive data of social network well;it has poor performance of complex network computing the basic parameters.The main purpose of this paper is to study how to use the new data storage technology,graphic database,to improve efficiency of data analysis of social network data.This paper takes the challenges of today's world of social network analysis,to explore some statistical parameters of social networks,with a certain degree of challenge.Firstly,social networks' concept is introduced and current research of social network data analysis is reviewed,also the relevant Neo4j graphic database processing engine,storage structure,access mode and Cypher query language are discussed.After studying Neo4j graphic database and it's currently used multidimensional index,which is lack of efficiency for querying large amounts of data,Skip List jump table index is proposed and applied to Neo4j graphic database to improve Neo4j query efficiency.Neo4j is a multidimensional index with index attribute in node or on the storage structure node or are linked list structure,retrieval traversal sequence.According to the list in a large amount of data,the query efficiency index began to decline.While SKIP LIST jump table index for Neo4j node or map database the relationship between the single linked list storage structure improved for the list of hierarchical indexing structure,the equivalent of two search,which greatly reduces the retrieval time and greatly improve the retrieval efficiency.then,based on the social network in some has represent the meaning of statistical features were analyzed,such as degree distribution,average path length,density,and other information are the cypher query language,and gives the realization of code.The Sina micro-blog' data is crawled as testing data set.Based on these data,the statistical characteristics of social network are analyzed,mainly includes the establishment of Neo4j graphic database nodes and relations,calculation of statistical characteristics,compared SKIP LIST index query with multidimensional index query and conventional Neo4j data.Experimental results show that,compared with the multidimensional index,the Neo4j graphic database Skip LIST jump table index is greatly improved in efficiency of analysis of the social network,and has a better user experience.The Neo4j graphic database improvement has value for the social network analysis.