João M. M. Duarte

A genetic algorithm-based approach to cost-sensitive bankruptcy prediction

The prediction of bankruptcy is of significant importance with the present-day increase of bankrupt companies. In the practical applications, the cost of misclassification is worthy of consideration in the modeling in order to make accurate and desirable decisions. An effective prediction system requires the integration of the cost preference into the construction and optimization of prediction models. This paper presents an evolutionary approach for optimizing simultaneously the complexity and the weights of learning vector quantization network under the symmetric cost preference. Experimental evidences on a real-world data set demonstrate the proposed algorithm leads to significant reduction of features without the degradation of prediction capability.

Typical load profiles in the smart grid context — A clustering methods comparison

The present research paper presents five different clustering methods to identify typical load profiles of medium voltage (MV) electricity consumers. These methods are intended to be used in a smart grid environment to extract useful knowledge about customers behaviour. The obtained knowledge can be used to support a decision tool, not only for utilities but also for consumers. Load profiles can be used by the utilities to identify the aspects that cause system load peaks and enable the development of specific contracts with their customers. The framework presented throughout the paper consists in several steps, namely the pre-processing data phase, clustering algorithms application and the evaluation of the quality of the partition, which is supported by cluster validity indices. The process ends with the analysis of the discovered knowledge. To validate the proposed framework, a case study with a real database of 208 MV consumers is used.

Learning manifolds for bankruptcy analysis

We apply manifold learning to a real data set of distressed and healthy companies for proper geometric tunning of similarity data points and visualization. While Isomap algorithm is often used in unsupervised learning our approach combines this algorithm with information of class labels for bankruptcy prediction. We compare prediction results with classifiers such as Support Vector Machines (SVM), Relevance Vector Machines (RVM) and the simple k-Nearest Neighbor (KNN) in the same data set and we show comparable accuracy of the proposed approach.

A stable credit rating model based on learning vector quantization

Credit rating is involved in many financial applications to estimate the creditworthiness of corporations or individuals. In addition to building accurate credit rating models, the stability of models is of significant importance to economic performance. In this work we propose a methodology based on learning vector quantization (LVQ) to develop a credit rating model. This model is applied to a French database of private companies over a period of several years. LVQ is trained and calibrated in a supervised way using data from 2006 and then applied to the remaining years. We analyze one year transition matrix and show that the model is capable to create robust and stable classes to rank companies.

Cost-Sensitive Learning Vector Quantization for Financial Distress Prediction

Financial distress prediction is of crucial importance in credit risk analysis with the increasing competition and complexity of credit industry. Although a variety of methods have been applied in this field, there are still some problems remained. The accurate and sensitive prediction in presence of unequal misclassification costs is an important one. Learning vector quantization (LVQ) is a powerful tool to solve financial distress prediction problem as a classification task. In this paper, a cost-sensitive version of LVQ is proposed which incorporates the cost information in the model. Experiments on two real data sets show the proposed approach is effective to improve the predictive capability in cost-sensitive situation.

Accurate prediction of financial distress of companies with machine learning algorithms

Prediction of financial distress of companies is analyzed with several machine learning approaches. We used Diane, a large database containing financial records from small and medium size French companies, from the year of 2002 up to 2007. It is shown that inclusion of historical data, up to 3 years priori to the analysis, increases the prediction accuracy and that Support Vector Machines are the most accurate predictor.

Cost-sensitive LVQ for bankruptcy prediction: An empirical study

Cost-sensitive learning is of critical importance in many domains including bankruptcy prediction where the costs of different errors are unequal. Most existing classification methods aim to minimize overall error based on the assumption that the costs are equal. This paper presents three cost-sensitive learning vector quantization (LVQ) approaches to incorporate cost matrix in classification. Experimental results on real-world data indicate the proposed approaches are effective alternatives for bankruptcy prediction in cost-sensitive situations.

Weighted learning vector quantization to cost-sensitive learning

The importance of cost-sensitive learning becomes crucial when the costs of misclassifications are quite different. Many evidences have demonstrated that a cost-sensitive predictive model is more desirable in practical applications than a traditional one without taking the cost into consideration. In this paper, we propose two approaches which incorporate the cost matrix into original learning vector quantization by means of instance weighting. Empirical results show that the proposed algorithms are effective on both binary-class data and multi-class data.

Hybrid Genetic Algorithm and Learning Vector Quantization Modeling for Cost-Sensitive Bankruptcy Prediction

Cost-sensitive classification algorithms that enable effective prediction, where the costs of misclassification can be very different, are crucial to creditors and auditors in credit risk analysis. Learning vector quantization (LVQ) is a powerful tool to solve bankruptcy prediction problem as a classification task. The genetic algorithm (GA) is applied widely in conjunction with artificial intelligent methods. The hybridization of genetic algorithm with existing classification algorithms is well illustrated in the field of bankruptcy prediction. In this paper, a hybrid GA and LVQ approach is proposed to minimize the expected misclassified cost under the asymmetric cost preference. Experiments on real-life French private company data show the proposed approach helps to improve the predictive performance in asymmetric cost setup.

Feature Selection for Bankruptcy Prediction: A Multi-Objective Optimization Approach

In this work a Multi-Objective Evolutionary Algorithm (MOEA) was applied for feature selection in the problem of bankruptcy prediction. This algorithm maximizes the accuracy of the classifier while keeping the number of features low. A two-objective problem, that is minimization of the number of features and accuracy maximization, was fully analyzed using the Logistic Regression (LR) and Support Vector Machines (SVM) classifiers. Simultaneously, the parameters required by both classifiers were also optimized, and the validity of the methodology proposed was tested using a database containing financial statements of 1200 medium sized private French companies. Based on extensive tests, it is shown that MOEA is an efficient feature selection approach. Best results were obtained when both the accuracy and the classifiers parameters are optimized. The proposed method can provide useful information for decision makers in characterizing the financial health of a company.