High-dimensional feature screening method based on hybrid ant colony optimization algorithm

Abstract

The invention discloses a high-dimensional feature screening method based on a hybrid ant colony optimization algorithm, mainly solving the problems of poor performance of a feature subset screened out in the prior art and long time consumption in the screening process. The technical scheme of the high-dimensional feature screening method includes the steps: preprocessing input high-dimensional data, and calculating the correlation and the symmetry uncertainty of the preprocessed data; initializing an ant colony algorithm, and constructing a feature subset through each ant; calculating the fitness of each feature subset, sorting the feature subsets, and selecting the contemporary optimal feature subset and the fitness thereof; iteratively updating the optimal feature subset of each generation and the fitness of the optimal feature subset; and comparing the fitness of the optimal feature subset of each generation, and taking the feature subset with the highest fitness as a final screened feature. According to the high-dimensional feature screening method, searching of the optimal feature subset is accelerated, and the screened feature subset has good classification performance, andthe overall operation time of screening is shortened, and the high-dimensional feature screening method can be used for channel selection in cancer gene analysis and hyperspectral image classification.

Description

technical field [0001] The invention belongs to the technical field of data processing, and in particular relates to a feature screening method, which can be used for channel selection in cancer gene analysis and hyperspectral image classification. Background technique [0002] Pattern recognition is a fundamental technique of artificial intelligence aimed at classifying objects into classes or categories. Recently, with the ease of data acquisition and the accumulation of large amounts of data, the main challenge in pattern recognition, namely the curse of dimensionality, has become more prominent. If the features are not screened and the data is used directly such as classification, the redundant features will increase the learning and testing time, and the irrelevant features will cause the performance of the classifier to decline. Especially in high-dimensional data sets, these shortcomings will be more obvious. Therefore, feature screening is necessary in data preproc...

AI Technical Summary

Problems solved by technology

However, most of the existing feature screening methods based on ant colony algorithm are applied to low-dimensional feature screening. If they are directly applied to high-dimensional feature screening, the running time of the algorithm will increase, and the performance of the obtained feature subset is also lower than that of traditional feature screening methods.

Images