Unlock instant, AI-driven research and patent intelligence for your innovation.

Feature selection method based on filter and RF-RFE algorithm

A technology of RF-RFE and feature selection method, applied in the field of feature selection based on filter and RF-RFE algorithm, to achieve the effect of high recognition ability

Inactive Publication Date: 2021-09-07
HARBIN UNIV OF SCI & TECH
View PDF0 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] In order to solve the problem of data set feature selection, the present invention discloses a feature selection method based on filter and RF-RFE algorithm

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Feature selection method based on filter and RF-RFE algorithm
  • Feature selection method based on filter and RF-RFE algorithm
  • Feature selection method based on filter and RF-RFE algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0044] In order to describe the technical solutions in the embodiments of the present invention, the present invention will be described in detail below in connection with the drawings in the examples.

[0045] Character selection with the ADNIMERGE data set of the Andi database as an example, such as figure 1 As shown, embodiments of the present invention provide a feature selection method based on a filter and an RF-RFE algorithm, including the following steps:

[0046] Step 1: Data pre-processing module, the complete data set is defective, discrete treatment, specifically:

[0047] Step 1-1 Filter the individual features for the original obtained data, setting the unreasonable value of the filter to null value, and the sample containing a lack value is deleted;

[0048]Step 1-2 After the screening, the AdniMege data set has a total of 21 features, and the target variable is divided into three categories, respectively, respectively, the old dementia crowd (AD), mild cognitive im...

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

No PUM Login to View More

Abstract

The invention relates to a feature selection method based on a filter and an RF-RFE algorithm, and the method comprises the following steps: a data preprocessing module carries out the missing value processing and discretization processing of a complete data set; the redundant feature processing module deletes redundant features by using a filter method combining information gain and correlation; the feature selection module performs feature selection on remaining features after redundant feature processing by adopting an RF-RFE method. According to the invention, the redundant features are deleted by combining correlation and information gain in a filter method, and the importance of variables is measured by adopting a recursive feature elimination (RFE) method and combining a random forest; the RF-RFE has high recognition capability in the aspect of searching feature subsets, a competitive result can be generated without adjusting parameters, and the high efficiency of feature selection is considered while the redundancy between the features is considered.

Description

Technical field: [0001] The present invention relates to the techniques of data classification, and more particularly to a feature selection method based on a filter and an RF-RFE algorithm, which has a good application in feature selection. Background technique: [0002] Feature Selection also said feature subset selection, is from the original feature to select some of the most effective features to reduce data set dimensions, and is an important means to improve learning algorithms. It is currently used for character selection methods. Three categories of filters, wrappers and embedded methods. [0003] FilterMethods is the most commonly used feature selection method, typically for single variable, it is assumed that each feature is independent of other features, the most famous filter method includes card square inspection, correlation coefficient, and information gain indicators. However, this filtering method leads to the loss of the related features, in order to overcome t...

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
no application Login to View More
IPC IPC(8): G06F16/9035G06F16/906G06K9/62
CPCG06F16/9035G06F16/906G06F18/2431
Inventor 苗世迪胡晓慧程可李静
Owner HARBIN UNIV OF SCI & TECH