A Hybrid Fourier Kernel Support Vector Machine Text Classification Method
A support vector machine and text classification technology, applied in the field of hybrid Fourier kernel function support vector machine text classification, can solve the problem of low accuracy, and achieve the effect of improving the effect, improving the performance, and the best text classification effect.
Active Publication Date: 2020-04-07
NANJING UNIV OF POSTS & TELECOMM +1
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The article [J.A.K.Suykens, J.Vandewalle, Least squares support vectormachine classifiers, Neural Processing Letters 9(3), 293(1999).] proposed the least squares support vector machine to solve nonlinear problems, but the accuracy is not very high
Literature [Zhang Yong. Performance analysis of Fourier kernel in support vector machine [D]. East China Normal University. 2008.] studied N-dimensional Fourier kernel on the basis of one-dimensional Fourier kernel, but the experimental analysis showed that On the text classification problem, the classification effect of N-dimensional and one-dimensional Fourier kernel functions is similar
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0048] Below in conjunction with the accompanying drawings and simulation results, a hybrid Fourier kernel function support vector machine text classification method proposed by the present invention will be described in detail:
[0049]A hybrid Fourier kernel function support vector machine text classification method, the implementation process of which is as follows:
[0050] Train a support vector machine to get α i and b, according to the Lagrangian multiplication and KKT conditions commonly used in optimization problems, the solution expressions are combined with equality constraints and inequality constraints, respectively, to simplify the support vector machine solution process, and the solution is transformed into:
[0051]
[0052] Restrictions: where C represents the slack variable;
[0053] In the formula, Represents the equivalent conversion result of the maximum interval of the support vector;
[0054] Indicates to find the minimum value of an expressio...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention proposes a mixed Fourier kernel function support vector machine text classification method. According to the different learning and generalization abilities of various kernel functions in the support vector machine, the method further forms a new mixed Fourier kernel function by linearly weighting the mixed polynomial and Fourier kernel functions; due to the learning ability of the kernel function and The generalization ability greatly affects the classification effect of the support vector machine, so the polynomial kernel function is combined with the Fourier kernel function. The method of the present invention inherits the high learning ability of the Fourier kernel function and the generalization ability of the polynomial kernel function, improves the performance of the support vector machine classifier; Compared with the polynomial in the kernel function and the mixed kernel function and the Gaussian kernel combination kernel function, the mixed Fourier kernel function has better generalization and learning ability, and the text classification effect is the best.
Description
technical field [0001] The invention is mainly applied to natural language processing in machine learning, and particularly relates to a hybrid Fourier kernel function support vector machine text classification method. Background technique [0002] With the advent of the era of big data, natural language processing, image processing and other related data processing have developed rapidly. Due to the high-dimensional characteristics of text information, how to find unique rules in these complex high-dimensional characteristics so as to serve people better in the future is an important research direction of statistical learning theory. Support Vector Machines (SVM) is a machine learning method based on statistical learning theory proposed by Vapnik et al in 1995. SVM solves nonlinear problems by relying on multiple kernel functions. [0003] At present, SVM has also been widely studied in nonlinear text classification problems. The article [Liu Gaohui, Yang Xing. A Support...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & AuthorityPatents(China)
IPC IPC(8): G06F16/35G06K9/62
CPCG06F18/2411
Inventor于舒娟张昀朱文峰何伟董茜茜金海红
OwnerNANJING UNIV OF POSTS & TELECOMM