Method and device for selecting central point of spectral clustering algorithm

The invention discloses a method and a device for selecting a central point of a spectral clustering algorithm. The method comprises the following steps: generating a step length by adopting a Levy flight strategy; judging whether the step length is greater than a preset threshold; if yes, updating the speed of each particle in the population by adopting a preset first formula; if not, a preset second formula is adopted to update the speed of each particle in the population, and the first formula and the second formula are not introduced into the step length; determining the position of each particle according to the updated speed of each particle in the population; for each particle in the population, substituting the position of the particle into a preset target function, and calculatingto obtain a fitness value of the particle; judging whether a termination condition is met currently or not; If yes, determining the particle with the optimal fitness value in the population as the center point of the spectra