A hypergraph key node detection method based on an exponential model

Abstract

The application discloses a kind of hypergraph key node detection method based on exponential model, belong to the field of identification hypergraph key node, including the following steps: hyperedge set represents hypergraph structure;Using s line chart converts hyperedge set into vertex set;Calculate the k Laplace energy;Calculate the difference of Laplace energy after deleting node;Calculate the Laplace energy centrality LC;Calculate the shortest average distance;Calculate the exponential centrality HEC of hypergraph;Calculate the top 5% node of real world hypergraph dataset;Using SIR model test method performance;Draw the curve of infection spread;Calculate the HEC of ten epileptic patients' epileptogenic zone and non-epileptogenic zone;Draw the contrastive graph of epileptogenic zone and non-epileptogenic zone HEC;Statistics HEC average and calculate p value;Draw distribution box line chart.The application combines the concept of exponential model and s line chart, can be used to identify key nodes in hypergraph.The application has greater improvement in identification accuracy and computational complexity compared with previous methods, and has good performance.

Images