Based on Randomly Executed Sequence and GA-SVC Applied on Air Quality Analysis

Abstract

Air quality is a key factor affecting people’s daily travel. In order to analyze and classify air quality, and to solve the problem of low classification accuracy, this paper mainly studied the SVC (multi-classification support vector machine) Algorithm based on the random sequence selection of optimized parameters through Genetic Algorithm. Aiming at the problems of insufficient classification accuracy in current multi-feature sequence analysis and unstable parameter selection in SVC algorithm, comprehensive classification evaluation of data is carried out by analyzing data features and correlation among features and integrating genetic algorithm into SVC to optimize parameter selection, so as to improve classification accuracy. The experimental results show that the classification accuracy has been improved by 5% on average with the current popular decision tree classification algorithm, unoptimized SVC algorithm and KNN algorithm.