Mobile application sorting and clustering method based on heterogeneous information network

Abstract

The present invention discloses a mobile application sorting and clustering method based on a heterogeneous information network. A sorting result mainly reflects the importance degree of an object, and thus, when the sorting result is introduced in the clustering process, a clustering result is more meaningful; moreover, the sorting result and the clustering result are continuously regulated by adopting an iterative method and supplement each other, so that a clustering effect is integrally promoted. Conventionally, only one or two types of information are used commonly in methods capable of being used for clustering of mobile applications; the mobile application sorting and clustering method is based on the heterogeneous information network consisting of four types of information of the applications and uses more information sources, so that clustering accuracy can be essentially promoted.

Description

technical field [0001] The invention belongs to the field of application recommendation, and in particular relates to a clustering method based on characteristics of heterogeneous information networks and sorting, and realizes a method for effectively clustering and sorting mobile applications. Background technique [0002] With the rapid development of the mobile Internet, a huge number of applications have emerged in the mobile market, and these various mobile applications are gradually changing people's lives. Each mobile application is associated with its own relevant information, and thousands of mobile applications form a huge heterogeneous information network, which contains a lot of valuable information, so the research on mobile application information network will is of great significance. On the one hand, in-depth analysis of the usage of a large number of mobile applications can help us understand the usage behavior of users in detail, so as to provide users wit...

AI Technical Summary

Problems solved by technology

At present, most of the methods that can be used for application data clustering analysis are mainly aimed at isomorphic information networks, that is, based on a certain type of application information, using a single type of information source ignores other relevant information to a large extent. limits the accuracy of clustering

Images