|
PresentationsSocial Network Graph Distributed Generation MethodDemidov Yaroslavl State University, Saltikova-Schedrina str., 59-12, Yaroslavl, 150014, Russia, e-mail: belov45@yandex.ru We plan to create social network graph clustering method. It has become necessary to generate a graph similar in structure to the existing social networks for future method testing. The article presents an algorithm for the graph distributed generation. Taken into account the social networks basic properties such as a power-law distribution of the users communities number dense crossing user communities and others. This algorithm has solved the problems that are present in the similar works of other authors. For example it is the multiple edges problem in the generation process. A created algorithm feature is implementation depending from communities number parameter rather than from users number as it is done in other works. It connects with existing social networks structure growth feature. There are properties of its graph in the paper. We describe a table containing the variables needed for the algorithm. By-step generation algorithm was compiled. Appropriate mathematical parameters was calculated for it. Generation performed distributed by Apache Spark framework. Described in detail how the tasks division with the help of this framework runs. There is Erdos-Renyi model for random graphs in the algorithm. It is the most suitable and easy to implement. The main advantages of created method are small resources amount in comparison with other similar generators and execution speed. Speed is achieved through distributed work and that in any time with algorithm distributed work network users have their own unique numbers and order by these numbers, so no need to sort them out. Designed algorithm will promote not only to the efficient clustering method creation. It can be useful in other development areas connected for example with the social networks search engines.
|
