The reliability and durability of battery management systems (BMS) in electric vehicles to predict the state of charge (SoC) is a difficult task. This paper presents terminal voltage estimation of lithium-ion battery systems using a machine learning algorithm to be applied in electric vehicles. The algorithms that can be used are linear regression, artificial neural network, Gaussian process regression, support vector machine, ensemble boosting, and ensemble bagging. In this paper, multivariable linear regression is used and applied to three battery data sets, and their accuracies are compared. The SoC estimation of the battery is vital for the BMS in electric vehicles. The result shows the accuracy of the machine learning model depends on the ratio in which the data sets are split and at a particular ratio the models attain the maximum accuracy.