A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
Prediction of the compressive strength of normal concrete using ensemble machine learning approach
https://orcid.org/http://orcid.org/0000-0002-2576-0482
In recent times, the use of machine learning techniques has become a new trend for predicting compressive strength of concrete. The use of such algorithms in predicting the strength and its proportioning has eased the manual work of mix design. In the present study, an attempt has been made to use advanced machine learning algorithms in predicting concrete strength. It has been performed using ensemble machine learning methods such as random forest regressor and Categorical Boosting (CatBoost) algorithm to predict the compressive strength of concrete. To reduce the underfitting or overfitting issue, a fivefold cross-validation technique has been used. In addition, a comparison has been made between both the algorithms considering performance indices, accuracy, and data importance. Furthermore, by considering different algorithms, the feature importance of each variable and the relationships between variables have been established for making the models more efficient. The analysis of the study exhibits that the random forest and CatBoost have a prediction accuracy of 89.3 and 95.8%, respectively. Thus, considering the accuracy and performance indices, CatBoost model has outperformed random forest model. Finally, this study demonstrates the possibilities of making the infrastructure sustainable and predictable, by analysing the mechanical properties of concrete using different machine learning algorithms.
Prediction of the compressive strength of normal concrete using ensemble machine learning approach
https://orcid.org/http://orcid.org/0000-0002-2576-0482
In recent times, the use of machine learning techniques has become a new trend for predicting compressive strength of concrete. The use of such algorithms in predicting the strength and its proportioning has eased the manual work of mix design. In the present study, an attempt has been made to use advanced machine learning algorithms in predicting concrete strength. It has been performed using ensemble machine learning methods such as random forest regressor and Categorical Boosting (CatBoost) algorithm to predict the compressive strength of concrete. To reduce the underfitting or overfitting issue, a fivefold cross-validation technique has been used. In addition, a comparison has been made between both the algorithms considering performance indices, accuracy, and data importance. Furthermore, by considering different algorithms, the feature importance of each variable and the relationships between variables have been established for making the models more efficient. The analysis of the study exhibits that the random forest and CatBoost have a prediction accuracy of 89.3 and 95.8%, respectively. Thus, considering the accuracy and performance indices, CatBoost model has outperformed random forest model. Finally, this study demonstrates the possibilities of making the infrastructure sustainable and predictable, by analysing the mechanical properties of concrete using different machine learning algorithms.
Prediction of the compressive strength of normal concrete using ensemble machine learning approach
https://orcid.org/http://orcid.org/0000-0002-2576-0482
In recent times, the use of machine learning techniques has become a new trend for predicting compressive strength of concrete. The use of such algorithms in predicting the strength and its proportioning has eased the manual work of mix design. In the present study, an attempt has been made to use advanced machine learning algorithms in predicting concrete strength. It has been performed using ensemble machine learning methods such as random forest regressor and Categorical Boosting (CatBoost) algorithm to predict the compressive strength of concrete. To reduce the underfitting or overfitting issue, a fivefold cross-validation technique has been used. In addition, a comparison has been made between both the algorithms considering performance indices, accuracy, and data importance. Furthermore, by considering different algorithms, the feature importance of each variable and the relationships between variables have been established for making the models more efficient. The analysis of the study exhibits that the random forest and CatBoost have a prediction accuracy of 89.3 and 95.8%, respectively. Thus, considering the accuracy and performance indices, CatBoost model has outperformed random forest model. Finally, this study demonstrates the possibilities of making the infrastructure sustainable and predictable, by analysing the mechanical properties of concrete using different machine learning algorithms.
Prediction of the compressive strength of normal concrete using ensemble machine learning approach
Asian J Civ Eng
Sapkota, Sanjog Chhetri (author) / Saha, Prasenjit (author) / Das, Sourav (author) / Meesaraganda, L. V. Prasad (author)
Asian Journal of Civil Engineering ; 25 ; 583-596
2024-01-01
14 pages
Article (Journal)
Electronic Resource
English
Prediction of the compressive strength of normal concrete using ensemble machine learning approach
| Springer Verlag | 2024
Environmentally Friendly Concrete Compressive Strength Prediction Using Hybrid Machine Learning
| DOAJ | 2022
Compressive Strength Prediction of Lightweight Concrete: Machine Learning Models
| DOAJ | 2022