Search

Evaluation of earthquake-induced liquefaction potential is crucial in the design phase of construction projects. Although several machine learning models achieve good prediction accuracy on their particular datasets, they may not perform well in other liquefaction datasets. To address this issue, we proposed a novel hybrid classifier ensemble to improve generalizability by combining the predictions of seven base classifiers using the weighted voting method. The applied base classifiers include back propagation neural network, support vector machine, decision tree, k-nearest neighbours, logistic regression, multiple linear regression and naïve Bayes. The hyperparameters and weights of the base classifiers were tuned using the genetic algorithm. To verify the robustness of the classifier ensemble, its performance was tested on three datasets collected from previous published researches. The results show that the proposed classifier ensemble outperforms the base classifiers in terms of a variety of performance metrics including accuracy, Kappa, precision, recall, F1 score, AUC and ROC on the three datasets. In addition, the importance of influencing variables was achieved by the classifier ensemble on the three datasets to facilitate the future data collecting work. This robust ensemble method can be extended to solve other classification problems in civil engineering. [ABSTRACT FROM AUTHOR]

Summary: Rockbust is a violent expulsion of rock due to the extreme release of strain energy stored in surrounding rock mass, leading to considerable damages to underground strucures and equipment, and threatening workers' safety. As the operational depth of engineering projects increases, a larger number of factors influence the mechanism of rockburst. Therefore, accurate classification of rockburst intensity cannot be achieved based on conventional criteria. It is urgent to develop new models with high accuracy and ease to implement in practice. This study proposed an ensemble machine learning method by aggregating seven individual classifiers including back propagation neural network, support vector machine, decision tree, k‐nearest neighbours, logistic regression, multiple linear regression and Naïve Bayes. In addition, we proposed nine data imputation methods to replace the missing values in the compiled database including 188 rockburst instances. Five‐fold cross validation and the beetle antennae search algorithm are used to tune hyperparameters and voting weights of the individual classifiers. The results show that the rockburst classification accuracy obtained by the classifier ensemble has increased by 15.4% compared with the best individual classifier on the test set. The predictor importance obtained by the classifier ensemble shows that the elastic energy index is the most sensitive input variable for rockburst intensity classification. This robust ensemble method can be extended to solve other classification problems in underground engineering projects. [ABSTRACT FROM AUTHOR]

• The UCS of SFC are predicted using BPNN models. • MOBAS is developed for multi-objective mixture optimization of SFC. • A GUI is developed for mixture optimization of SFC. Partial replacement of cement by silica fume in concrete provides advantages such as mitigation of the impact on the environment of carbon dioxide emitted during cement production, recycling of industrial by-products and improvement of concrete strength and durability. The optimization of the mixture of silica fume concrete (SFC) requires trade-off among multiple objectives (strength, cost and embodied CO 2) and consideration of a large number of variables under highly nonlinear constraints. Obtaining the Pareto front of this multi-objective optimization (MOO) problem is computationally expensive. To address this issue, the present study develops a MOO model using machine learning (ML) techniques and a new meta-heuristic algorithm. Firstly, the relationships between components and SFC properties are modelled on a dataset using a back propagation neural network (BPNN) model. Then an individual-intelligence-based multi-objective beetle antennae search algorithm (MOBAS) is developed to search for optimal SFC mixtures that maximize UCS, and minimize cost and embodied CO 2 under defined constraints. Results indicate that the proposed MOBAS is more computationally efficient with satisfactory accuracy in comparison with algorithms based on swarm intelligence. The MOO model achieves reliable predictions for UCS with a very high correlation coefficient (0.9663) on the test set. The Pareto front of optimal SFC mixture proportions of the MOO problem is successfully obtained using the proposed model. The proposed frame-work improves the efficiency in SFC mixture optimization and can facilitate appropriate decision making before construction. [ABSTRACT FROM AUTHOR]