Visualization Classification Method of Multi-dimensional Data Based on Fourier Transform
Wenxue Hong · Jisuanji gongcheng · 2008
A novel method for designing the classifier of multi-dimensional data is proposed,which uses radar chart of multi-statistics to show multidimensional data and applies Fourier descriptors to recognize the radar chart.Different multi-dimensional data forms different radar chart and distinguishes different category.And Fourier descriptor takes as chart characteristic,by Fourier transform on the boundary curve of radar chart.Based on the improved Probabilistic Neural network(PNN),radar chart is recognised automatically.Experimental results show this method has better result of classification,classification precision is out approximately 8.25 percentage,compared with the traditional classifier.