Mobile application homology edge clustering method based on multi-dimensional features

The invention discloses a mobile application homology edge clustering method based on multi-dimensional features. The mobile application homology edge clustering method comprises the following steps:constructing a mobile application software relational graph, taking mobile application software as a vertex of the relational graph, obtaining n multi-dimensional feature similarity scores between software through different feature comparison algorithms of the mobile application software and as edges of the relational graph, and forming an edge set; setting a similar score threshold value and an effective score number threshold value, and filtering out an effective edge set; calculating the distance between every two effective edges; setting a neighborhood radius, and finding out neighborhoodsof all effective edges; setting a neighborhood density threshold value, finding out all core edges, and forming a core edge object set; and traversing from any core edge in the core edge object set,clustering all effective ed