Estimation of shear strength parameters of soil using Optimized Inference Intelligence System

In recent years, machine learning techniques have been developed and used to build intelligent information systems for solving problems in various fields. this study, we Optimized Inference Intelligence System namely ANFIS-PSO which is a combination of Adaptive Neural-Fuzzy (ANFIS) Particle Swarm Optimization (PSO) the estimation shear strength parameters soils (Cohesion “C” angle internal friction “φ”). These are required designing foundation civil engineering structures. Normally, soil determined either field or laboratory require time, expertise equipments. Therefore, applied hybrid model quick cost-effective based on other six physical clay content, natural water specific gravity, void ratio, liquid limit plastic limit. data 1252 soft samples collected from different highway project sites Vietnam. The was randomly divided into 70:30 ratios training testing, respectively. Standard statistical measures: Root Mean Square Error (RMSE), Absolute (MAE) Correlation Coefficient (R) were performance evaluation model. Results study indicated that very good predicting soil: cohesion (RMSE = 0.075, MAE 0.041, R 0.831) 0.08, 0.058, 0.952).