A Review of LSSVM Mathematical Models and Applications

Least Squares Support Vector Machines (LSSVM) is an extension algorithm of the standard Support Vector Machines (SVM), which takes the least squares system as the loss function and transforms the original quadratic programming problem into a linear problem for solving, which greatly reduces the computational complexity. The complexity of operation is greatly reduced, and the topic has emerged as a focal point of research in the realm of machine learning. This paper reviews the theoretical foundation and research status of least squares support vector machine, and analyzes in detail the application of LSSVM in the fields of fault diagnosis, network intrusion, and medical treatment.