Twin Support Vector Regression Machine And Modified Algorithm Based On L₁ Norm

This paper mainly studies the comparison of the advantages and disadvantages ofL₁-norm under different twin support vector regression machines.This paper mainly divided into four parts,the first chapter introduced the SVR models used in this paper.The second chapter introduced the application ofL₁-norm on support vector regression machine,gives the corresponding learning algorithms,and gives their advantages.Finally,the feasibility and competitiveness of the algorithm are verified by experiments.In the third chapter,the algorithm ofL₁-e-TSVR model is introduced,and gives the corresponding learning algorithms.In the fourth chapter,the application ofL₁-norm on modified support vector regression machine is introduced,and gives the corresponding learning algorithms.Finally,the effectiveness of the learning algorithm is verified by experiments.The advantages of the algorithm in computing speed and accuracy are proved.