Research on the Massive Redundant Data Mining Algorithm based on Kernel Clustering and Data Cleaning Technology

In this research article, we conduct theoretical research on massive redundant data mining algorithm based on kernel clustering and the data cleaning technology. Clustering is an important application in the field of data mining and it based on the correlation between samples will be divided into several categories and samples of the similar within the same category, for different methods and different types of sample. We propose the kernel based methodology as the condition that directly on the sample characteristics of clustering, but clustering effect is largely dependent on the distribution of sample points when all kinds of boundary of the sample linear time sharing has good performance. Our research leads the new perspective of data analysis and mining, in the future, more modification on the data cleaning and clustering will be proposed.