2017 IEEE/ACIS 16th International Conference on Computer and Information Science (ICIS) (2017)
May 24, 2017 to May 26, 2017
Yongchang Wang , Communication University of China, Beijing, China
Ligu Zhu , Communication University of China, Beijing, China
With the arrival of the era of big data, people's ability to collect and obtain data is becoming more powerful. These data have shown the characteristics of high dimension, large scale and complex structure. High dimensional data has seriously hindered the efficiency of data mining algorithm, we call it "the Dimension disaster ". Therefore, dimension reduction technology has become the primary task of big data mining and machine learning. In this paper, we focus on the method of data reduction, described the category of data dimension reduction. The research status and main algorithms of dimension reduction method are described in detail. This paper briefly introduces the latest research progress of data dimension reduction algorithm, including some popular algorithm such as PCA, KPCA, SVD, etc. The principle of principal component analysis (PCA) is discussed in this article, and the singular value decomposition (SVD) theorem is introduced to solve the problem that the PCA method has a large amount of computation, we also give a comparison of PCA and SVD. Finally, we design and implement some experiments to verify the application of SVD in data analysis and latent semantic indexing.
Principal component analysis, Matrix decomposition, Eigenvalues and eigenfunctions, Singular value decomposition, Data mining, Algorithm design and analysis, Feature extraction
Y. Wang and L. Zhu, "Research and implementation of SVD in machine learning," 2017 IEEE/ACIS 16th International Conference on Computer and Information Science (ICIS), Wuhan, China, 2017, pp. 471-475.