Earth pressure balanced (EPB) shield machines are large and complex mechanical systems and have been widely applied to tunnel projects. The earth pressure in the chamber is very important for EPB shield machines to avoid ground settlement and guarantee safe construction during the tunnelling process. In order to improve the prediction accuracy of the earth in the chamber, we propose a new earth pressure prediction method in this paper. At first, we analyse the main factors influencing the earth pressure in the chamber and determine the inputs and output of the prediction model. Then, we use the least squares support vector machine to establish the prediction model whose parameters are optimised by the particle swarm optimisation. The experimental results demonstrate that the proposed method has high prediction accuracy and is an effective way to predict the earth pressure in the chamber for EPB shield machines.