Analyzing Customer Relationships In Social Networks Of Small Groups Using Frequent Subgraph Discovery

With the rapid development of information technology,the Internet enters the web2.0 era.In this era,Internet has undergone major changes,and social networking has become a typical application of this era.Inside online social networks(OSNS),users comment on certain topics,share ideas,and create relationships.With most users spending a significant amount of time on them,OSNS have seen a dramatic rise in popularity and owned growth in the number.Different social networks hope to stand out in a number of similar products,so a critical aspect for OSNS success is how efficient their recommendation mechanisms are in modeling user behavior and providing suggestions tailored to users' needs.In this paper,research point is relationship of small group social network between social network users,which mainly includes:1.In view of the above requirements,we use nodes to represent individual entities and use edges to represent the relationship or interaction between the nodes in this paper.And based on frequent subgraph discovery to achieve the social network analysis and predict the relationships among social network users,which including "ego-graph construction module","frequent subgraph discovery module" and "graph data vectorization and classification module",the result achieve visualization by relationship graph.2.Graph mining technologies and their algorithms are studied in order to choose a rational algorithm to each module function.And three strategies are proposed to against the shortcomings of gSpan for high complexity,inefficient and the selection of k.In"frequent subgraph discovery module",the frequent graph discovery is realized based on the improved gSpan algorithm,and show that such patterns possess enough discriminative power to accurately predict the relationships among social network users.3.We evaluate the approach through an experimental study that comprises large-scale,real-world datasets and show that the relationships between users are strongly associated with the frequent subgraph around them,at the same time,we evaluate the validity and reliability of the method.