Research on algorithms for multimodal knowledge fusion in intelligent knowledge graph construction

With the rapid development of digital technology and intelligent information systems, the application of multimodal knowledge fusion in the construction of intelligent knowledge graphs has attracted more and more attention. However, the research on the construction of intelligent knowledge graphs is still relatively limited and needs to be further explored. As an important part of the field of artificial intelligence, the complex internal relationships and multi-source heterogeneous data environment of intelligent knowledge graphs pose new challenges to researchers and developers. Therefore, this paper proposes to build a more intelligent and efficient knowledge graph based on the wavelet-fractional multivariate grey prediction model and combines digital technology and intelligent information systems. Wavelet analysis enables us to better understand the different scale characteristics in multimodal data, while fractional calculus provides a powerful mathematical tool for dealing with nonlinear and non-stationary problems. The support of digital technology and intelligent information systems makes the parameter estimation and training process of the model more accurate, and improves the ability to capture complex relationships in the knowledge graph. Through the wavelet - fractional multivariate grey prediction model, multimodal data can be analyzed more comprehensively, thereby providing a scientific basis and technical support for the construction of intelligent knowledge graphs.