A High-dimensional data analysis method based on PCA and density clustering in clinical diagnosis

With the development of medical science and technology, a large number of high-dimensional data have appeared in medical fields. There is a large amount of irrelevant data and data redundancy information, which leads to the learning model tends to transition, fitting or characterization ability weakened. It is unable to express effectively the medical data characteristics. By principal component analysis and principal component analysis based on mutual information, we select features and reduce dimensionality of high dimensional data. Furthermore, combine with effective density clustering with excellent performance to reduce ultimately the size of data and improve the quality of data analysis to help doctors making diagnosis. Experimental results show that this method can effectively deal with high-dimensional medical data analysis and achieve better results.