A Sequential Structure for Water Inflow Forecasting in Coal Mines Integrating Feature Selection and Multi-Objective Optimization

The accurate classification of reservoir recovery factor is dampened by irregularities such as noisy and high-dimensional features associated with the reservoir measurements or characterization. These irregularities, especially a larger number of features, make it difficult to perform accurate classification of reservoir recovery factor, as the generated reservoir features are usually heterogeneous. Consequently, it is imperative to select relevant reservoir features while preserving or amplifying reservoir recovery accuracy. This phenomenon can be treated as a multi-objective optimization problem, since there are two conflicting objectives: minimizing the number of measurements and preserving high recovery classification accuracy. In this study, wrapper-based multi-objective feature selection approaches are proposed to estimate the set of Pareto optimal solutions that represents the optimum trade-off between these two objectives. Specifically, three multi-objective optimization algorithms—Non-dominated Sorting Genetic Algorithm II (NSGA-II), Multi-Objective Grey Wolf Optimizer (MOGWO) and Multi-Objective Particle Swarm Optimization (MOPSO)—are investigated in selecting relevant features from the reservoir dataset. To the best of our knowledge, this is the first time multi-objective optimization has been used for reservoir recovery factor classification. The Artificial Neural Network (ANN) classification algorithm is used to evaluate the selected reservoir features. Findings from the experimental results show that the proposed MOGWO-ANN outperforms the other two approaches (MOPSO and NSGA-II) in terms of producing non-dominated solutions with a small subset of features and reduced classification error rate.

Download Full-text

Comparison of Profit-Based Multi-Objective Approaches for Feature Selection in Credit Scoring

Algorithms ◽

10.3390/a14090260 ◽

2021 ◽

Vol 14 (9) ◽

pp. 260

Author(s):

Naomi Simumba ◽

Suguru Okami ◽

Akira Kodaka ◽

Naohiko Kohtake

Keyword(s):

Feature Selection ◽

Selection Process ◽

Credit Scoring ◽

Computational Time ◽

Multi Objective Optimization ◽

Multi Objective ◽

Base Classifier ◽

Grasshopper Optimization Algorithm ◽

Specific Implementation ◽

The Impact

Feature selection is crucial to the credit-scoring process, allowing for the removal of irrelevant variables with low predictive power. Conventional credit-scoring techniques treat this as a separate process wherein features are selected based on improving a single statistical measure, such as accuracy; however, recent research has focused on meaningful business parameters such as profit. More than one factor may be important to the selection process, making multi-objective optimization methods a necessity. However, the comparative performance of multi-objective methods has been known to vary depending on the test problem and specific implementation. This research employed a recent hybrid non-dominated sorting binary Grasshopper Optimization Algorithm and compared its performance on multi-objective feature selection for credit scoring to that of two popular benchmark algorithms in this space. Further comparison is made to determine the impact of changing the profit-maximizing base classifiers on algorithm performance. Experiments demonstrate that, of the base classifiers used, the neural network classifier improved the profit-based measure and minimized the mean number of features in the population the most. Additionally, the NSBGOA algorithm gave relatively smaller hypervolumes and increased computational time across all base classifiers, while giving the highest mean objective values for the solutions. It is clear that the base classifier has a significant impact on the results of multi-objective optimization. Therefore, careful consideration should be made of the base classifier to use in the scenarios.

Download Full-text