Graph Embedding Using Frequency Filtering

The target of graph embedding is to embed graphs in vector space such that the embedded feature vectors follow differences and similarities source graphs. In this paper, a novel method named Frequency Filtering Embedding (FFE) proposed which uses Fourier transform filtering as domain operator for extraction. amplifies or attenuates selected frequencies using appropriate filter functions. Here, heat, anti-heat, part-sine identity sets are A generalized version FFE GeFFE also by defining pseudo-Fourier operators. This can be considered general framework formulating some previously defined invariants other works choosing suitable bank flexibility empowers adapt itself properties each dataset unlike previous spectral methods leads superior classification accuracy relative others. Utilizing set, its members different parts spectrum turn, improves method. Additionally, resolves cospectrality problem entirely tested datasets.