Prediction and Stability Assessment of Soft Foundation Settlement of the Fishbone-Shaped Dike Near the Estuary of the Yangtze River Using Machine Learning Methods

Fishbone-shaped dikes are always built on the soft soil submerged in water, and foundation settlement plays a key role stability of these dikes. In this paper, novel simple approach was proposed to predict fishbone by using extreme learning machine. The machine is single-hidden-layer feedforward network with high regression classification prediction accuracy. data-driven models were based small training sample size fast speed. simulation results showed that methods had good performances facilitating comparisons measured data predicted data. Furthermore, final dike models, fishbone-shaped assessed model. findings paper suggested method could be an effective tool for assessment