A meta-path-based anomaly motif query method in heterogeneous networks

Abstract

The invention provides a meta-path-based abnormal motif query method in a heterogeneous network, belonging to the field of network abnormal query. This method is aimed at the data extraction network containing various types of information. By setting the initial conditions required for the query, the set of all motifs that meet the conditions is found, and the normalized path similarity is used to compare each motif with the reference model. The similarity between the motif sets, for the motifs with low similarity, we consider them to be abnormal motifs in the target motif. Experimental results show that the method proposed by the invention can be applied to a variety of heterogeneous networks and achieve good results. Since there are fewer methods for querying abnormal motifs in heterogeneous networks, the invention can be used as a The abnormal motif query method is used stably, thus providing a new perspective for abnormal motif query.

Description

Technical field [0001] The present invention belongs to the field of network abnormal query, mainly applied to querying the abnormal small industries included in the target mandarin collection in the heterogeneous network. Mainly calculated by calculating the number of gear paths between the body as the similarity and normalization using the mold to its own element path, the similarity between the target mandrel between the network and the intended mold is obtained, and the analysis is determined. The least similar part of the body is judged in a given network. Background technique [0002] With the arrival of the big data age and the development of the science and technology increases significantly, many data that need to consume a large amount of human material, and many data that are difficult to simply pass the data perspective, and now you can efficiently use your computer. Accurate analysis, which makes we can liberate from the complicated physical labor, more focusing on t...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of the lack of effective means for querying abnormal motifs in the network existing in the existing research. The number of meta-paths between a certain type of standard motif specified by the user is used to measure the similarity between the target motif and the standard motif, and the number of meta-paths from the motif to itself is used as the standard to normalize the similarity. The results serve as a basis for measuring relative similarity

Images