A Comparative Study on Feature Selection in Text Categorization
ICML '97 Proceedings of the Fourteenth International Conference on Machine Learning
Efficient and Effective Clustering Methods for Spatial Data Mining
VLDB '94 Proceedings of the 20th International Conference on Very Large Data Bases
Laplacian Eigenmaps for dimensionality reduction and data representation
Neural Computation
A Projection Pursuit Algorithm for Exploratory Data Analysis
IEEE Transactions on Computers
Applied Multidimensional Scaling
Applied Multidimensional Scaling
| Hi-index | 0.00 |
Research on existing vector representation method of the text data and it is advantages or disadvantages and applicability of linear and nonlinear dimensionality reduction methods, the paper analysis the feature of laws and regulations of construction data, establishes the vector space model of the text data. Based on Riemann Surface Theory and Topology Theory, adopting the globe pole mapping method of space analytic geometry, forming a homeomorphism with the linear analytic function from the high-dimensional vector to the low one, and then realizing a corresponding mapping, maintained the topological structure of data before and after dimensionality reduction, at last, the results solve the neighbor problem of nonlinear dimensionality reduction for text mining effectively. Appling the method in the text data of laws and regulations of construction, it has obtained the projection in any two-or-three dimensional orthogonal and univocal direction, caparisoning with other research methods shows the validity of this method.