Hybrid Leapfrog Algorithm in Feature Selection Optimization of Text Classification

Abstract

The research of hybrid leapfrog algorithm has always been a hot spot. The performance of the classifier model can be improved by improving and optimizing the model. This paper proposes a fusion text semantic structure pattern recognition system based on hybrid leapfrog and support vector machine. The leapfrog algorithm is studied to improve the speed and accuracy of text classification feature selection. This paper mainly uses experimental design and data comparison to illustrate the performance of different algorithms in text classification. The experimental results show that MSFLA-FCM algorithm has great advantages in terms of average fitness value, the maximum fitness reaches 1.518, and the effect of text classification is significantly improved.