Rui Xiao-Ping
College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China
Dong Cheng-Wei
Beijing Institute of Surveying and Mapping, Beijing, 100038, China
Song Xian-feng
College of Resources and Environment, University of Chinese Academy of Sciences, Beijing, 100049, China
ABSTRACT
Manifold learning is a nonlinear dimension reduction method that is increasingly capturing researchers interests. It can be used to discover the intrinsic structure of sample data. This study discusses the application of the Locally Linear Embedding (LLE) method on dimensionality reduction for sparse and non-uniform economic statistics data. Aiming at the sparse and uneven distribution of the characteristics of the statistical sample data, this study develops an Adaptive LLE (ALLE) method and adopts a new distance metric and neighbor value selection method for calculating the neighbor points. The improved methods are employed in an experiment that analyzes the 2007 Sichuan statistical data from China. By visualizing the dimension reduction data, the experiment shows that the improved method can retain more information and reveal the distribution of regional economic situation more accurately when a suitable nearest neighbor value is selected. By comparing the results from ALLE when different nearest neighbor values are used, it is shown that the classifications still depend on the selected neighbors. Finally, this paper explains the results calculated using different target dimensions and shows that the results can reflect the economic status of the Sichuan district better when the intrinsic dimensionality is selected optimally.
PDF References Citation
How to cite this article
Rui Xiao-Ping, Dong Cheng-Wei and Song Xian-feng, 2013. Improved Locally Linear Embedding Method Suitable for Multi-dimensional Visualization of Economic Statistics. Information Technology Journal, 12: 8651-8656.
DOI: 10.3923/itj.2013.8651.8656
URL: https://scialert.net/abstract/?doi=itj.2013.8651.8656
DOI: 10.3923/itj.2013.8651.8656
URL: https://scialert.net/abstract/?doi=itj.2013.8651.8656
REFERENCES
- Belkin, M. and P. Niyogi, 2003. Laplacian eigenmaps for dimensionality reduction and data representation. Neural Comput., 15: 1373-1396.
CrossRef - Parmanto, B., M. Paramita, W. Sugiantara, G. Pramana, M. Scotch and D. Burke, 2008. Spatial and multidimensional visualization of Indonesia's village health statistics. Int. J. Health Geographics, Vol. 7.
CrossRef - Roweis, S.T. and L.K. Saul, 2000. Nonlinear dimensionality reduction by locally linear embedding. Science, 290: 2323-2326.
CrossRefDirect Link - Tenenbaum, J.B., V. de Silva and J.C. Langford, 2000. A global geometric framework for nonlinear dimensionality reduction. Science, 290: 2319-2323.
CrossRefDirect Link - Zhang, Z. and H. Zha, 2005. Principal manifolds and nonlinear dimensionality reduction via tangent space alignment. SIAM J. Sci. Comput., 26: 313-338.
CrossRef