Search

A metaheuristic-based structural optimization method, whilst being popularly adopted due to its advantages in by-passing gradient function calculations, requires the use of time-consuming advanced analyses for constraint evaluation. To overcome this drawback, the present paper proposes a robust (machine learning-based) optimization method that combines the light gradient boosting machine (LightGBM) with the efficient p-best differential evolution (E p DE) method. In essence, the LightGBM classification model is constructed to assess the constraint (safety and integrity) satisfaction of structures. An efficient framework using a so-called safety parameter is proposed to prevent inaccurate predictions of the LightGBM model. The E p DE processes the optimization procedures on the constructed classification LightGBM model. This enables an enhanced machine learning-based optimization technique that not only maintains the sufficiently accurate optimal design of structures but also significantly reduces the required computing efforts, as compared to standard optimization schemes. Various examples of steel structure designs (i.e., two of which have been provided herein) have been successfully performed by the proposed approach. These illustrate the accuracy and robustness of the proposed method, where good comparisons with reference algorithms (including standard DE with "DE/rand/1" mutational strategy, Jaya, Rao-1 and CaDE) are evidenced. The statistical values collected present the accuracy and reliability of the proposed method in obtaining the minimum total weight designs with substantial (some 60%) reduction of computing efforts required to furnish optimal results. [ABSTRACT FROM AUTHOR]

• A three-phase hybrid sampling method is proposed to handle class imbalance problem. • HVDM distance is used to determine the type of examples. • Imbalance ratio is embedded to the formula to categorize the majority class examples. • Over-sampling phase is similar to Safe-level-smote and LN-smote. • The safe level ratio will be used as the power of two to generate random numbers. In two-class classification problems, learning from imbalanced data is a challenging task due to the bias of machine learning algorithms towards the majority class. It has been shown in some studies that the imbalance problem is not the only reason for this difficulty, and other issues related to the nature of the data, such as small disjunct problems, borderline and rare examples, overlap between classes, and data shift, also contribute. To address this issue, an Under-Sampled Weighted AVeraGed BorderlineSMOTE (USWAVG-BS) is proposed in this paper, which consists of three phases: determining the types of examples, under-sampling, and over-sampling. In the first phase, the feature space is transformed into the Heterogeneous Value Distance Metric (HVDM) space, and a new formula based on the imbalance ratio is defined to identify noise examples from the majority class. In the second phase, noise examples from the majority class are either converted to the minority class or removed until a specific threshold between the two classes is reached. In the last phase, new examples are generated using a similar way to Safe-level-SMOTE and LN-SMOTE to further boost the minority class by emphasizing minority class regions. The proposed approach's performance is evaluated using recall, precision, and f1-score as performance criteria on 10 real-world data sets using the JRip algorithm. Moreover, USWAVG-BS is compared with other sampling methods including 3 under-sampling, 7 over-sampling, and 4 hybrid ones using f1-score metric. The obtained results indicate that the proposed approach outperforms other methods significantly using the Wilcoxon signed-rank test. It is worth mentioning that RandomUnderSampler, despite its simplicity, performs better than two other under-sampling methods on average and SMOTE performs better than its extensions such as BorderlineSMOTE and ASN-SMOTE on the examined data sets. [ABSTRACT FROM AUTHOR]

Classification tasks are of great importance in machine learning. However, class imbalance is a universal problem that needs to be solved in classification and can greatly affect the performance of machine learning classifiers. Developing from the basic belief rule base (BRB) system, the hierarchical belief rule-based system can integrate expert knowledge and has the potential to alleviate the negative effect of class imbalance. To utilize BRB to solve the imbalanced multi-classification task and avoid the combinational explosion problem, a novel hierarchical BRB structure based on the extreme gradient boosting (XGBoost) feature selection method, abbreviated as HFS-BRB is proposed in this paper in order to deal with any number of classes. In the hierarchical BRB structure, there is one main-BRB in the first level and several sub-BRBs in the second level. The XGBoost technique is used for feature selection in the modelling process of each abovementioned BRB model. The output of the main BRB represents the approximated classification between confusable classes. Then, these samples were transmitted to a certain sub-BRB for binary classification to make a precise prediction. Thus, a multi-classification problem can be transformed into several binary classification problems. The class imbalance is alleviated. To test the effectiveness of the proposed method, seven classical benchmark problems for imbalanced classification and a real asteroid orbit classification were performed. [ABSTRACT FROM AUTHOR]

Corporate financial distress prediction research has been ongoing for more than half a century, during which many models have emerged, among which ensemble learning algorithms are the most accurate. Most of the state-of-the-art methods of recent years are based on gradient boosted decision trees. However, most of them do not consider using feature importance for feature selection, and a few of them use the feature importance method with bias, which may not reflect the true importance of features. To solve this problem, a heuristic algorithm based on permutation importance (PIMP) is proposed to modify the biased feature importance measure in this paper. This method ranks and filters the features used by machine learning models, which not only improves accuracy but also makes the results more interpretable. Based on financial data from 4,167 listed companies in China between 2001 and 2019, the experiment shows that compared with using the random forest (RF) wrapper method alone, the bias in feature importance is indeed corrected by combining the PIMP method. After the redundant features are removed, the performance of most machine learning models is improved. The PIMP method is a promising addition to the existing financial distress prediction methods. Moreover, compared with traditional statistical learning models and other machine learning models, the proposed PIMP-XGBoost offers higher prediction accuracy and clearer interpretation, making it suitable for commercial use. • The model Combines a corrected feature selection measure and XGBoost. • Permutation importance can correct the bias of feature importance. • The model is validated on Chinese listed companies datasets over five metrics. • The model is proved to outperform several benchmark techniques. • The feature importance and partial dependence plot enhance model interpretation. [ABSTRACT FROM AUTHOR]

• Two-phase training algorithm able to boost the Machine Learning accuracy. • Accurate bankruptcy prediction up to 60 months by a Machine Learning algorithm. • Integration of the Machine Learning algorithm in a DSS for policy making. • Test of the DSS on the Italian SMEs, with a focus on the COVID-19 effects. In the field of detection and prediction of company defaults and bankruptcy, significant effort has been devoted to evaluating financial ratios as predictors using statistical models and machine learning techniques. This problem becomes crucially important when financial decision-makers are provided with predictions on which to act, based on the output of prediction models. However, research has shown that such predictors are sufficiently accurate in the short-term, with the focus mainly directed towards large and medium-large companies. In contrast, in this paper, we focus on mid- and long-term bankruptcy prediction (up to 60 months) targeting small and/or medium enterprises. The key contribution of this study is a substantial improvement of the prediction accuracy in the short-term (12 months) using machine learning techniques, compared to the state-of-the-art, while also making accurate mid- and long-term predictions (measure of the area under the ROC curve of 0.88 with a 60 month prediction horizon). Extensive computational tests on the entire set of companies in Italy highlight the efficiency and accuracy of the developed method, as well as demonstrating the possibility of using it as a tool for the development of strategies and policies for entire economic systems. Considering the recent COVID-19 pandemic, we show how our method can be used as a viable tool for large-scale policy-making. [ABSTRACT FROM AUTHOR]