Multiobjective optimization of support vector regression parameters by teaching-learning-based optimization for modeling of electric discharge machining responses

In this chapter, the development of a unified structure of support vector machine regression for predicting multiple responses is attempted. Unified learning is performed by simultaneous minimization of errors in estimation of material removal rate and average surface roughness by modified teaching–learning-based optimization (TLBO). This development is an advancement of mathematical modeling toward the compact virtual data generator. In the proposed modified TLBO, combined rank method, an improvement in multiobjective optimization by TLBO, is suggested for simultaneous optimization of multiple objective functions, and an optimum unique set of C, ε, and σ is obtained.