Abstract:Existing clustering algorithms have the problems of low precision and easy to fall into local optimum. The paper proposes a new algorithm—ISFLA-K, which combined with an improved shuffled frog leaping algorithm and K-Means clustering algorithm. The algorithm uses the idea of an independent study to generate the initial population. According to the frogs' characteristics of cognitive and learning ability, it improve the rules of shuffled frog leaping algorithm leapfrog. The paper uses ISFLA to optimize K-Means clustering algorithm, which improved solution accuracy. The experimental results can prove the validity and superiority of the proposed algorithm.