An Efficient Multi-layer Ensemble Framework with BPSOGSA-Based Feature Selection for Credit Scoring Data Analysis

Neighborhood rough set and SVM based hybrid credit scoring classifier

The detection of variables that contribute to the variation of a system is one of the most important considerations in the industrial manufacturing processes. This work presents the combination of Mahalanobis–Taguchi system and a hybrid binary metaheuristic based on particle swarm optimization and gravitational search algorithm (BPSOGSA) to perform an optimal feature selection in order to detect the relevant variables in a real process of foam injection in automotive industry. The proposed method is compared with other feature selection approach based in binary PSO algorithm. The experimental results revealed that BPSOGSA is faster and successfully converge selecting a smallest subset of features than BPSO. Moreover, the feature selection effect is validated through other widely used machine learning algorithms which improve their accuracy performance when they are trained with the subset of detected variables by the proposed system.

Credit scoring has become an important management science issue as the credit industry has been experiencing enormous growth during the past few decades. Numerous popular classification methods (e.g. linear discriminant analysis, quadratic discriminant analysis, and logistic regression) have been applied in credit scoring for years. Recently researchers proposed several sophisticated and highly effective data mining techniques, such as Skew-normal discriminant analysis (SNDA), Skew-t discriminant analysis (STDA), Stepwise discriminant analysis (SDA), Sparse discriminant analysis (Sparse DA), Flexible discriminant analysis (FDA), and Mixture discriminant analysis (MDA). The objective of this study is to examine these recently proposed discrimination methods for screening credit applicants. The performance of various credit scoring models is evaluated by one real-world credit scoring dataset. The predictive ability of each credit scoring model is accessed by the total percentage of correctly classified cases (total PCC) and the bad rate among accepts (BRA). The results show that SNDA, STDA, and SDA are outperforming techniques for implementing credit scoring models.

The adoption of machine learning (ML) in financial services has fundamentally transformed credit evaluation, loan approval, and risk prediction processes. Traditional statistical models, while interpretable and regulatory-compliant, are often inadequate for capturing the complex, non-linear relationships and dynamic patterns present in modern financial datasets. This chapter presents a comprehensive exploration of ML techniques applied to credit scoring and financial risk management, highlighting their advantages in predictive accuracy, operational efficiency, and inclusion of alternative data sources. Supervised and unsupervised learning models, ensemble frameworks, and deep learning architectures are examined for their effectiveness in classifying borrower creditworthiness, forecasting default probabilities, and identifying anomalous financial behaviors. Special emphasis was placed on handling challenges associated with imbalanced and high-dimensional datasets, ensuring robust model performance across diverse borrower populations. The chapter discusses automated underwriting systems, real-time decision-making, and strategies to enhance financial inclusion for underbanked and marginalized populations. Ethical, regulatory, and explainability considerations are addressed through the integration of explainable AI and fairness-aware methodologies, ensuring transparency, accountability, and compliance in predictive financial models. Scenario analysis and stress testing using predictive analytics are explored as key tools for proactive risk mitigation and portfolio management, enabling institutions to anticipate adverse conditions and maintain operational resilience. By bridging theoretical advancements and practical applications, this chapter provides a structured framework for the deployment of ML-driven financial decision-making systems, emphasizing accuracy, equity, and adaptability in an increasingly data-intensive financial landscape.

Credit scoring has become a challenging issue for stock brokerages, as the credit industry has been facing high competition during the past decade. Many methods have been suggested to credit scoring in the literature. The purpose of this study is to set up a hybrid model for customer credit scoring in Iran's National Investment Brokerage. It also provides a way to pay appropriate facilities as tools for CRM. So, after the data pre-processing step, we convert refined dataset into RFM model. Customers were clustered using two clustering algorithms, self-organising map (SOM) and K-means. In both methods, the best optimum number of clusters was calculated at ten. Afterwards, the clusters ranked using TOPSIS and the top three clusters were considered as the target customers. Eventually, the credit score of the superior cluster members were calculated. Coefficient facilities granted to the top three clusters respectively are 0.271, 0.173 and 0.556.

PurposeThis paper aims to present a literature review of the most recent optimisation methods applied to Credit Scoring Models (CSMs).Design/methodology/approachThe research methodology employed technical procedures based on bibliographic and exploratory analyses. A traditional investigation was carried out using the Scopus, ScienceDirect and Web of Science databases. The papers selection and classification took place in three steps considering only studies in English language and published in electronic journals (from 2008 to 2022). The investigation led up to the selection of 46 publications (10 presenting literature reviews and 36 proposing CSMs).FindingsThe findings showed that CSMs are usually formulated using Financial Analysis, Machine Learning, Statistical Techniques, Operational Research and Data Mining Algorithms. The main databases used by the researchers were banks and the University of California, Irvine. The analyses identified 48 methods used by CSMs, the main ones being: Logistic Regression (13%), Naive Bayes (10%) and Artificial Neural Networks (7%). The authors conclude that advances in credit score studies will require new hybrid approaches capable of integrating Big Data and Deep Learning algorithms into CSMs. These algorithms should have practical issues considered consider practical issues for improving the level of adaptation and performance demanded for the CSMs.Practical implicationsThe results of this study might provide considerable practical implications for the application of CSMs. As it was aimed to demonstrate the application of optimisation methods, it is highly considerable that legal and ethical issues should be better adapted to CSMs. It is also suggested improvement of studies focused on micro and small companies for sales in instalment plans and commercial credit through the improvement or new CSMs.Originality/valueThe economic reality surrounding credit granting has made risk management a complex decision-making issue increasingly supported by CSMs. Therefore, this paper satisfies an important gap in the literature to present an analysis of recent advances in optimisation methods applied to CSMs. The main contribution of this paper consists of presenting the evolution of the state of the art and future trends in studies aimed at proposing better CSMs.

Nature is a great source of inspiration for solving complex problems in real-world. In this paper, a hybrid nature-inspired algorithm is proposed for feature selection problem. Traditionally, the real-world datasets contain all kinds of features informative as well as non-informative. These features not only increase computational complexity of the underlying algorithm but also deteriorate its performance. Hence, there an urgent need of feature selection method that select an informative subset of features from high dimensional without compromising the performance of the underlying algorithm. In this paper, we select an informative subset of features and perform cluster analysis by employing a cross breed approach of binary particle swarm optimization (BPSO) and sine cosine algorithm (SCA) named as hybrid binary particle swarm optimization and sine cosine algorithm (HBPSOSCA). Here, we employ a V-shaped transfer function to compute the likelihood of changing position for all particles. First, the effectiveness of the proposed method is tested on ten benchmark test functions. Second, the HBPSOSCA is used for data clustering problem on seven real-life datasets taken from the UCI machine learning store and gene expression model selector. The performance of proposed method is tested in comparison to original BPSO, modified BPSO with chaotic inertia weight (C-BPSO), binary moth flame optimization algorithm, binary dragonfly algorithm, binary whale optimization algorithm, SCA, and binary artificial bee colony algorithm. The conducted analysis demonstrates that the proposed method HBPSOSCA attain better performance in comparison to the competitive methods in most of the cases.

One of the most important aspects of risk management for financial institutions is assessing credit risk. Credit scoring models are important tools for evaluating loan applications because they provide a systematic way to assess credit worthiness. While traditional statistical models have been widely used, artificial intelligence (AI) has emerged as a more efficient alternative due to its ability to process large datasets and improve predictive accuracy. The growing reliance on AI-powered models has transformed lending practices, improving decision-making, reducing default risks, and enhancing financial stability. The focus of this research is on exploring AI-based credit scoring models and their impact on financial institutions. Traditional credit scoring methods often lack accuracy and efficiency, leading to increased risks and losses. AI methods like machine learning and deep learning offer a more reliable method, analyzing huge amounts of data and spot patterns that people are not aware of. Gaining insight into how AI affects credit scoring helps with risk management, loan selection, and financial inclusion. Other options for A1, A2, A3, A4, and A5. Income level, credit score, existing debt, and recent credit inquiries are all part of the assessment. The results showed that A3 ranked lowest and A4 ranked best. A1 has the highest value for The Role of AI in Enhancing Credit Scoring Models for Better Lending according to the TOPSIS Method approach.

The integration of Artificial Intelligence (AI) in credit scoring has transformed financial decision-making, offering enhanced accuracy, efficiency, and scalability.AI-powered credit scoring models leverage vast datasets and sophisticated machine learning algorithms to predict creditworthiness, reducing reliance on traditional credit histories.This innovation expands financial access, particularly for underserved populations, by incorporating alternative data sources such as social behavior, utility payments, and mobile transactions.However, the deployment of AI-driven credit models raises significant ethical concerns, primarily related to algorithmic bias, transparency, and regulatory compliance.Bias in AI credit scoring can emerge from historical data imbalances, leading to discriminatory outcomes that disproportionately affect marginalized groups.Addressing these challenges necessitates bias reduction strategies, including algorithmic fairness techniques, rigorous model auditing, and diversified training data.Financial institutions must ensure interpretability and accountability in AI models to foster consumer trust and regulatory adherence.Furthermore, AI-powered credit scoring plays a pivotal role in advancing financial inclusion.By utilizing alternative credit assessment methodologies, AI-driven models enable fairer access to credit for individuals with limited or no formal financial history.Collaboration between financial institutions, policymakers, and technology providers is essential to establish ethical AI frameworks that mitigate risks while promoting responsible lending practices.Striking a balance between innovation and fairness in AI credit scoring is crucial for ensuring equitable financial opportunities.Future research should focus on refining ethical AI principles, developing robust bias mitigation techniques, and establishing standardized governance frameworks.This study underscores the significance of ethical AI adoption in credit scoring, advocating for transparent, inclusive, and responsible financial practices.

Credit scoring models are the cornerstone of the modern financial industry. After years of development, artificial intelligence and machine learning have led to the transformation of traditional credit scoring models based on statistics. In this study, a novel multi-stage ensemble model with a hybrid genetic algorithm is proposed to achieve accurate and stable credit prediction. To alleviate the adverse effects of imbalanced data in credit scoring models, the Instance Hardness Threshold method is extended using a majority voting strategy to deal with data imbalance. To eliminate redundant and irrelevant features in the dataset and select well-performing base classifiers, a new hybrid genetic algorithm is proposed to obtain the optimal feature subset and base classifier subset. To aggregate the predictive power of the base classifiers, a stacking approach is adopted to integrate the optimal base classifiers into the ensemble model. The proposed model is tested on three standard imbalanced credit scoring datasets, compared with similar state-of-the-art approaches, and evaluated using four well-known evaluation indicators. The experimental results prove the effectiveness of the proposed model and demonstrate its superiority.

ABSTRACTCurrent studies have designed many credit scoring models with high performance, but they are often weak in interpretability with obvious “black box” features. This makes them difficult to meet the requirements of the regulators about the model's interpretability. This paper presents a novel credit scoring model as the IWSL model, which is data feature driven with interpretable features. The IWSL model first calculates the representative eigenvectors of default and nondefault samples according to their spatial distribution characteristics, as well as the eigenvector located in the middle of these two types of eigenvectors in the sample space. It then calculates the weighted distance between each sample and each eigenvector to divide the training dataset into three subsets and constructs sublogistic models separately. In the absence of prior information about the optimal weight setting of each attribute, the swarm intelligence algorithm is applied to back‐optimize the weights according to the model's generalization ability in the validation stage. The empirical results show that the IWSL model outperforms 12 leading credit scoring models on three public consumer credit scoring datasets with statistical significance. Model component validity testing confirms the effectiveness of the IWSL model's core settings, while sensitivity analysis validates its ability to maintain robust results.

The adaptation of the radio parameters is one of the basic capabilities of cognitive radio decision engine. Many researches focus on using genetic algorithm (GA) to select the optimal parameters. However, the convergence speed of GA is low. Recently, particle swarm optimization (PSO) has been used a lot in the cognitive radio system. In this paper, we proposed a hill-climbing binary particle swarm optimization (BPSO) which optimize optimal individual after one iterative operation by hill-climbing algorithm. The proposed method would enhance the local search capability at the later stage of each generation of BPSO. We designed a multi-carrier system for performance analysis. Through different weighting scenarios multiple objective fitness functions, the simulation results illustrate the trade-off between the fitness function and the transmission parameters configuration. And the results show that the hill-climbing BPSO is better than pure BPSO in stability and average fitness value.

To date, the usage of electromyography (EMG) signals in myoelectric prosthetics allows patients to recover functional rehabilitation of their upper limbs. However, the increment in the number of EMG features has been shown to have a great impact on performance degradation. Therefore, feature selection is an essential step to enhance classification performance and reduce the complexity of the classifier. In this paper, a hybrid method, namely, binary particle swarm optimization differential evolution (BPSODE) was proposed to tackle feature selection problems in EMG signals classification. The performance of BPSODE was validated using the EMG signals of 10 healthy subjects acquired from a publicly accessible EMG database. First, discrete wavelet transform was applied to decompose the signals into wavelet coefficients. The features were then extracted from each coefficient and formed into the feature vector. Afterward, BPSODE was used to evaluate the most informative feature subset. To examine the effectiveness of the proposed method, four state-of-the-art feature selection methods were used for comparison. The parameters, including accuracy, feature selection ratio, precision, F-measure, and computation time were used for performance measurement. Our results showed that BPSODE was superior, in not only offering a high classification performance, but also in having the smallest feature size. From the empirical results, it can be inferred that BPSODE-based feature selection is useful for EMG signals classification.

Small and Medium Enterprises (SMEs) have played a significant role in the development of any economy. However, easy access to finance from financial institutions is a prime challenge for them. Similarly, financial institutions also face difficulties while selecting the potential SMEs for granting credit. The SMEs are often seen as unorganized in terms of financial data as compared to large corporate sectors. The credit risk assessment based on unorganized financial data is a challenge for financial institutions. Most of the existing models used regression to predict the possibility of default of SMEs. However, the regression model may not perform well with limited data points and missing data. The problem can be solved by using a multi criteria decision‐making (MCDM) model. Credit scoring, especially addressing the SMEs, has been infrequently reported in the archived literature. To fill the gaps of literature, the present study proposes a credit scoring model applying the hybrid analytic hierarchy process‐technique for order of preference by similarity to ideal solution (AHP‐TOPSIS) technique. The study has been carried out in three stages. In the first stage, credit rating criteria and sub‐criteria have been identified from the literature review and taking opinions from experts. In the second stage, weights of criteria and sub‐criteria have been calculated using AHP. Finally, in the third stage, weights calculated by AHP have been used in TOPSIS to determine the credit score. The effectiveness of the proposed model has been illustrated through a case study. Further, the results of the proposed model are compared with the commercially available ratings. The proposed model may be a low‐cost alternative for financial institutions for credit scoring of SMEs. Further, the model has the advantage of customization as per the needs of the financial institutions. The suggested model can help the managers to identify the potential SMEs for granting credit.

An artificial intelligence system for predicting customer default in e-commerce