Abstract
A high dimensional data visualization platform based on nonlinear dimension reduction approach was built and deployed in order to research the visualization of large-scale user linked feature data. The following test results were obtained through the implementation of a dimension reduction method and a functional module: The test set of the MNIST data set is given in CSV format, which may be represented as a 785*10000 matrix. The matrix is a representation of the handwritten picture that has been abstracted and transformed. The PCA approach provides the best dimensional-reduction impact on the dietary nutrient dataset, retaining 98.8 percent of the variation. The protein structure and function data set is not well served by any of the three dimensional-reduction techniques. Both T-SNE and Large Vis algorithms have better dimensional-reduction effects on MNIST data set, which reflects the nonlinear characteristics of the data set. Compared with T-SNE algorithm, Large Vis algorithm has no significant improvement in visualization effect, which is mainly reflected in time efficiency.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
