Multi-modal evolution feature automatic conformal representation method based on dynamic hypergraph network

Abstract

The invention belongs to the field of big data machine automatic learning, particularly relates to a multi-modal evolution feature automatic conformal representation method, system and apparatus based on a dynamic hypergraph network, and aims to solve the problem that existing multi-modal feature representation can only represent multi-modal data static features and is difficult to represent multi-modal high-order dynamic association features. The method comprises the following steps: acquiring data stream training samples of m modals as input data streams; extracting a finite node set of m modal feature vectors in the input data stream; generating a Laplacian matrix of the m modal hypergraphs; performing inter-hypergraph high-order correlation conformal entropy solution calculation of m modes to generate a multi-mode high-order dynamic correlation form alignment model of n topic network sets; and performing increase and decrease alignment on the newly added hypergraph vertexes and nodes covered by the multi-modal high-order dynamic association form alignment model to realize automatic updating of the topic network set. According to the method, the problem that multi-modal high-order big data multivariate correlation evolution characteristics are difficult to characterize is solved.

Description

technical field [0001] The invention belongs to the field of automatic learning of big data machines, and in particular relates to an automatic conformal representation method, system and equipment for multi-modal evolution features based on a dynamic hypergraph network. Background technique [0002] With the integration of information technologies such as the Internet, cloud computing, and artificial intelligence, and industries such as manufacturing, transportation, urban management, and medical care, the interaction between people and people, and between people and things is facilitated, and text, image, video, etc. The rapid expansion of information such as audio and audio, along with the Internet of Things such as radar, infrared, multimedia sensors, and infrastructure perception systems, has been continuously integrated into social computing systems, triggering rapid data growth, forming a multi-source heterogeneous, fast, Massive data collections that are difficult to...

AI Technical Summary

Problems solved by technology

[0005] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem that it is difficult to express multi-modal high-order dynamic associations in the existing multi-modal static feature representation, the first aspect of the present invention proposes a dynamic hypergraph based An automatic conformal representation method for multi-modal evolution features of the network, the method includes:

Images