A Multidimensional Data Feature Selection Method Combining Genetic Algorithm and Dragonfly Algorithm
A feature selection method and multi-dimensional data technology, applied in the field of machine learning, can solve problems such as premature convergence of genetic algorithm, feature combination is not the optimal solution, etc., and achieve the effect of reducing optimization time and optimizing optimization results
Active Publication Date: 2022-05-24
JILIN UNIV
View PDF3 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
The researchers applied the genetic algorithm to the feature selection process to obtain a better feature combination, but due to the premature convergence of the genetic algorithm, the feature combination may not be the optimal solution.
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
[0101] see figure 1 As shown, the steps of a multidimensional data feature selection method combining genetic algorithm and dragonfly algorithm are as follows:
[0102] 1) Simple cleaning of traffic accident data;
[0103] 2) using the data feature screening result in step 1) as the input of a multidimensional data feature selection method combining genetic algorithm and dragonfly algorithm, and the output result is the data feature selected by the method of the present invention;
[0104] see figure 2 As shown in Figure 1, the traffic accident data is simply cleaned, and according to each dimension of the data, the features that have only a single value in this dimension and the data is missing more than half are filtered out, and the information entropy value is ranked in the bottom three from large to small. Feature screening, that is, this feature is not selected when all data is used for model training, and the remaining features are reserved. The data screening steps...
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
A multi -dimensional data feature selection method combined with genetic algorithm and dragonfly algorithm. The specific steps are: simply clean the traffic accident data; embed the dragonfly algorithm into the genetic algorithmTo increase the speed of searching for the genetic algorithm; embed the dragonfly algorithm into the genetic algorithm to interfere with the mutant operation, whether the "food" and "natural enemy" gene position calculated through the dragonfly algorithm is selected, and set the probability of different genetic position mutation.To increase the convergence of algorithms.The data feature screening results are used as a multi -dimensional data feature selection method that combines the genetic algorithm and the dragonfly algorithm, and the output result is the data feature selected by the algorithm.Experiments have proved that this method has good performance for different classifiers, and verify that the characteristic selection method of the present invention is effective and has robustness.
Description
technical field [0001] The present invention relates to a method belonging to machine learning, more precisely, the present invention designs a multi-dimensional data feature selection method combining genetic algorithm and dragonfly algorithm. The present invention can not only be applied to the field of machine learning, but also can be extended to other fields, and other fields also belong to the protection scope of this patent. Background technique [0002] Machine learning is an emerging discipline in recent years, involving many fields. Machine learning specializes in how computers simulate or implement human learning behaviors to acquire new knowledge or skills and continuously improve their performance. In machine learning, the selection of data sets is particularly important. For the selection of data sets, the first is to select a data set with enough features and samples, and the second is to perform feature selection in the data set. Since optimization in the ...
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 & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/00G06N3/12
CPCG06N3/006G06N3/126G06F18/2415G06F18/214
Inventor 杨晓萍王星乔柳莹于树友李娟
Owner JILIN UNIV