Prediction of Superconducting Transition Temperatures for Fe- Based Superconductors using Support Vector Machine

Abstract

Quench for materials that can persistently carry electrical current without loss of power is confined to low temperatures. The future dream of room temperature superconductors is hampered by the absence of unique theory that fully explains superconductivity as well as high cost of the equipment involved in the characterization of the potential samples. Support vector machine (SVM) is hereby proposed to predict the superconducting transition temperature of any family of iron-based superconductors at ambient pressure using lattice parameters of the samples. Accuracy of over 99% obtained in our developed model is not only creating an efficient and low cost way of predicting transition temperature but also  makes lattice parameter a premise through which full understanding of superconductivity can be grown.

Keywords: Iron-based superconductor, Support vector machine, correlation coefficient and superconducting transition temperature