An ensemble method to improve prediction of earthquake-induced soil liquefaction: a multi-dataset study

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 naive 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.

Automated summary

A novel hybrid classifier ensemble method was proposed to improve the generalizability of earthquake-induced liquefaction potential evaluation models by combining the predictions of seven base classifiers using weighted voting. The ensemble method outperformed the base classifiers in terms of various performance metrics on three datasets, and also identified the importance of influencing variables for future data collection. This robust method can be extended to solve classification problems in civil engineering.