Research on Support Vector Machine in Automobile Evaluation and Decision Making

This study systematically evaluates the application value of machine learning algorithms in automobile evaluation and decision making. By comparing the performance of mainstream algorithms, support vector machine (SVM) is determined as the benchmark model, and quantum particle swarm optimization (QPSO) algorithm is introduced to achieve parameter adaptive optimization. The experimental results show that the SVM model optimized by QPSO is significantly superior to the traditional grid search optimization method in terms of classification performance indexes, and the accuracy (98.55% vs. 95.67%) and accuracy (98.60% vs. 96.23%) are improved by 2.88% and 2.37%, respectively. The superiority of the proposed method in automobile evaluation and decision making task is verified.