Learning decision boundaries for cone penetration test classification

In geotechnical field investigations, cone penetration tests (CPT) are increasingly used for ground characterization of fine-grained soils. Test results different parameters that typically visualized in CPT based data interpretation charts. this paper we propose a novel methodology which is on supervised machine learning permits redefinition the boundaries within these charts to account unique soil conditions. We train ensembles randomly generated artificial neural networks classify six types database hundreds from Austria and Norway. After training combine multiple solutions classification problem visualize new decision between types. The comprehensible step towards automatically adjusted specific local