Predictive Modeling of Concrete Compressive Strength using Machine Learning Algorithms

The culture and modern-day society reflect their attributes with their style of architecture. A particular style holds value and certain characteristics. In today's world, concrete is a common factor across nations and cultures. It has proven to be a fundamental component of construction. It has so far become the most used man-made material. Concrete is made by mixing aggregates like cement, water, sand, gravel, and other strength-boosting materials, and later hydration is the process that converts the moldable mixture into a solid. Statistical metrics such as mean absolute error, mean squared error, root mean square error, and coefficient of determination are used to test the learning models. To create ensemble models, several techniques were examined and used, including support vector regression, random forest, and k-nearest neighbors. using a coefficient of determination of 0.98, stacking-based regression using support vector regression, random forest, and k-nearest neighbor as the final estimator beat other models in the study, including bagging and random forest. Furthermore, a web application was created using the trained machine learning models for user-friendliness.