Analysis of mathematical modeling in particular clustering process of mixed data

the analysis method of mathematical modeling in particular clustering process of mixed data is of great significance for improving the ability of data analysis. The traditional method for specific clustering process mathematics modeling of mixed data is based on K-Means clustering algorithm, it is easy to fall into local convergence, and clustering effect is poor. Therefore, the analysis method of mathematical modeling in particular clustering process of mixed data is proposed in the paper based on particle swarm density maximum distance concave function and boundary membership degree feature analysis. The mixed data clustering sample points are divided into k classes according to the degree of similarity to cluster centers, dimensionality reduction is performed for differentiation characteristics of primitive variable data, through searching particles in the space, each particle has the speed, position and fitness, and the optimal solution is found by iteration, preprocessing for data standardization is conducted, data preprocessing includes scale selection of number, type and characteristics, boundary membership feature analysis is processed to achieve mathematical modeling analysis for specific clustering process of mixed data. The simulation results show that, the algorithm has the superior clustering performance of mixed data, good convergence, and great application value.