A tumor identification method based on deterministic particle swarm optimization and support vector machine

Abstract

The invention discloses a tumor identification method based on deterministic particle swarm optimization and support vector machine, which includes preprocessing the tumor gene expression spectrum data, using the classification information index method to preliminarily select the information gene on the training set, and then using The pairwise redundancy method removes redundant genes to obtain the candidate gene pool; further uses the classification information index method to obtain key gene subsets on the training set; uses the deterministic particle swarm optimization algorithm on the training set to optimize the parameters of the support vector machine Optimize, and then identify the gene expression profile data of the tumor to be identified. The invention makes full use of the feature that the support vector machine is suitable for small-sample data identification, uses deterministic particle swarm optimization to optimize the support vector machine, further improves the performance of the support vector machine, and thus improves the accuracy of tumor identification.

Description

technical field [0001] The invention belongs to the application field of computer analysis technology of tumor gene expression spectrum data, and in particular relates to a tumor identification method based on deterministic particle swarm optimization and support vector machine. Background technique [0002] DNA microarray technology has brought great opportunities to biology, but the massive and complex microarray data it generates has posed great challenges to scholars in related fields. There are four main reasons for this: first, microarray data contains a lot of noise or outliers. Because noise and outliers are often generated during the experiment, and the data processing process will also bring errors or sample category labeling errors, it is hoped that a robust processing method can be designed. Second, gene expression profile data is huge, and how to deal with large-scale data sets is also one of the difficulties that needs to be solved. Therefore, it is very mean...

AI Technical Summary

Problems solved by technology

[0002] DNA microarray technology has brought great opportunities to biology, but the massive and complex microarray data it generates has posed great challenges to scholars in related fields. There are four main reasons for this: first, microarray data contains a lot of noise or outliers

Because noise and outliers are often generated during the experiment, and the data processing process will also bring errors or sample category label errors, therefore, it is hoped that a robust processing method can be designed

Second, the scale of gene expression profile data is huge, and how to deal with large-scale data sets is also one of the difficulties that need to be solved

However, due to the randomness of the particle search, the traditional PSO blindly searches more times, resulting in a longer search time, and the search performance needs to be improved.

Images