Data filling method and device based on ant lion algorithm, equipment and storage medium

Abstract

The embodiment of the invention relates to the technical field of data processing, in particular to a data filling method and device based on an ant lion algorithm, equipment and a storage medium, and aims to improve the data filling quality in a missing data filling task. The method comprises the following steps: receiving a missing data set, setting a plurality of ant individuals and ant lion individuals according to missing data types, calculating fitness values of each ant individual and each ant lion individual, selecting the ant lion individual with the maximum fitness value as an elite ant lion individual, selecting one selected ant lion individual for each ant individual, and selecting the selected ant lion individual as the elite ant lion individual; updating the position codes of the ant individuals according to the position codes of the selected ant lion individuals and the position codes of the elite ant lion individuals, updating the selected ant lion individuals according to the updated ant individuals so as to determine new elite ant lion individuals, iteratively executing the steps, generating a data filling model according to the finally obtained elite ant lion individuals, and completing the data filling. And executing the data filling task.

Description

technical field [0001] The embodiments of the present application relate to the technical field of data processing, and in particular, relate to a data filling method, device, device and storage medium based on the antlion algorithm. Background technique [0002] With the rapid development of the network, a large amount of data will be generated on the network every day. In order to extract valuable information from the massive data on the network, data mining technology emerges as the times require. To mine data, you need to use a classifier, and the classifier needs to be trained through a high-quality data set. If the quality of the training data is poor, it will affect the performance of the classifier, and then affect the effect of data mining. useless information and ignore useful information. Due to some unavoidable technical problems, when obtaining the data set for training, there will inevitably be over-loss in the stages of data collection and storage, and the da...

AI Technical Summary

Problems solved by technology

To mine data, you need to use a classifier, and the classifier needs to be trained through a high-quality data set. If the quality of the training data is poor, it will affect the performance of the classifier, and then affect the effect of data mining. Useless information, ignore useful information

Due to some unavoidable technical problems, when obtaining the data set for training, there will inevitably be over-loss in the stages of data collection and storage, and the data in the data set is missing. If the missing data is directly ignored, it may cause Potentially valuable information is missing, so the data set needs to be filled with missing data

[0003] In the existing technology, the distance-based method is used for data filling, which does not distinguish the characteristics of different data types, and does not make full use of the distribution and probability information of the characteristics. As a result, the data tends to be consistent and loses diversity, and the filled data will also appear different from When the deviation between actual data is too large, the trained classifier has a poor effect on data mining and cannot meet the data quality requirements of existing data mining tasks.

Images