Visualization Of P2P Lending Data Based On Parallel Coordinates And Data Mining

In this paper we present an approach studying real network P2 P lending financial data, using data visualization tools along with some algorithms of data mining to find out the factors which affect the success rate of a lending transaction. A variety of algorithms and manifestations in the field of Computational Aesthetics and Data Mining are being used in this work.This work utilizes parallel coordinates as a means of visualization method and makes some important improvements. The concept Data Density is promoted here, in order to display the results more effectively and ensure the efficiency at the same time. This paper uses a number of interactive tools to achieve more flexible results. Animation is also included in the methods here.Kmeans algorithm is used for clustering P2 P lending data and presenting more significant results. Using a combination of data mining algorithms with heuristic algorithms, the initial data and cluster parameters for Kmeans algorithm are being modified. In the ending stage, this modified algorithm is fused into parallel coordinates displaying method, showing a concept which is familiar to people and intuitively clear.