Combining PSO cluster and nonlinear mapping algorithm to perform clustering performance analysis: take the enterprise financial alarming as example

Algorithm by simulating biological intelligence concept with application in the optimization issue is still in the emergent stage. Among them, Particle Swarm Optimization Algorithm is a concept and method that has group intelligence. Using the exploring and development feature of particle swarm, the best solution of the entire domain is searched in the question space. However, Clustering Analysis has an objective to plan the data set according to certain principle into several sub-sets. Due to the practicability of Clustering Analysis, many scholars propose different clustering algorithms to be used by the researchers. Therefore, in this article, Particle Swarm Optimization Algorithm is adopted to self-write Clustering Analysis program. Then Fuzzy Sammon Mapping nonlinear mapping algorithm is associated to perform clustering performance and classificaton capability assessment. From the test result of test data of enterprise’s financial alarming, it can be seen that Particle Swarm Clustering Analysis can obtain good clustering performance, and good classification capability can be obtained too.