A Corporate Credit Rating Model Using Support Vector Domain Combined with Fuzzy Clustering Algorithm

Guo, Xuesong; Zhu, Zhengwei; Shi, Jia

doi:10.1155/2012/302624

Cited by 11 publications

(9 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Different multiclass methods do not influence the performance of threshold-based algorithms to a great extend. Nevertheless, for ANNs and SVMs, there is a big performance drop when using the one-vs-rest method which was observed in other studies [ 48,49,76] as well. The remaining multiclass methods only show marginal differences among each other.…”

Section: Performance Of All Models and Methodsmentioning

confidence: 65%

“…Guo et al [49] studied credit rating with four classes as well. They used a support vector domain combined with a fuzzy clustering algorithm and compared it with different SVM multiclass methods.…”

Section: Machine Learning In Finance For Multiclass Credit Ratingmentioning

confidence: 99%

“…RFs are used to represent combined interpretable models using thresholds. We compare the multiclass methods used by Guo et al [49] and an ensemble method representatively for the work of Kwon et al [50]. Furthermore, we use our new approach to express doubt in the classification.…”

Section: Interpretable Models In Financementioning

confidence: 99%

See 2 more Smart Citations

Interpretable Multiclass Models for Corporate Credit Rating Capable of Expressing Doubt

Obermann

Waack

2016

Front. Appl. Math. Stat.

View full text Add to dashboard Cite

Corporate credit rating is a process to classify commercial enterprises based on their creditworthiness. Machine learning algorithms can construct classification models, but in general they do not tend to be 100% accurate. Since they can be used as decision support for experts, interpretable models are desirable. Unfortunately, interpretable models are provided by only few machine learners. Furthermore, credit rating often is a multiclass problem with more than two rating classes. Due to this fact, multiclass classification is often achieved via meta-algorithms using multiple binary learners. However, most state-of-the-art meta-algorithms destroy the interpretability of binary models. In this study, we present Thresholder, a binary interpretable threshold-based disjunctive normal form (DNF) learning algorithm in addition to modifications of popular multiclass meta-algorithms which maintain the interpretability of our binary classifier. Furthermore, we present an approach to express doubt in the decision of our model. Performance and model size are compared with other interpretable approaches for learning DNFs (RIPPER) and decision trees (C4.5) as well as non-interpretable models like random forests, artificial neural networks, and support vector machines. We evaluate their performances on three real-life data sets divided into three rating classes. In this case study all threshold-based and interpretable models perform equally well and significantly better than other methods. Our new Thresholder algorithm builds the smallest models while its performance is as good as the best methods of our case study. Furthermore, Thresholder marks many potential misclassifications in advance with a doubt label without increasing the classification error.

show abstract

Section: Performance Of All Models and Methodsmentioning

confidence: 65%

Section: Machine Learning In Finance For Multiclass Credit Ratingmentioning

confidence: 99%

See 1 more Smart Citation

Interpretable Multiclass Models for Corporate Credit Rating Capable of Expressing Doubt

Obermann

Waack

2016

Front. Appl. Math. Stat.

View full text Add to dashboard Cite

show abstract

“…The ratings of the agencies that sort risk are viewed by financial agents as efficient for risk management, although they represent a very costly procedure [3]. However, recent events have tarnished the image of these agencies.…”

Section: Introductionmentioning

confidence: 99%

Sovereign Credit Risk Assessment with Multiple Criteria Using an Outranking Method

Silva

et al. 2018

Mathematical Problems in Engineering

View full text Add to dashboard Cite

In view of the records of failures in rating agencies’ assessments for sorting countries’ quality of credit in degrees of default risk, this paper proposes a multicriteria sorting model using reference alternatives so as to allocate sovereign credit securities into three categories of risk. From a numerical application, what was observed from the results was a strong adherence of the model in relation to those of the agencies: Standard & Poor's and Moody's. Since the procedure used by the agencies is extremely subjective and often questioned, the contribution of this paper is to put forward the use of an objective and transparent methodology to sort these securities. Given the agencies’ conditions for undertaking the assessment, a complete similarity between the results obtained and the assignments of the agencies was not expected. Therefore, this difference arises from subjective factors that the agencies consider but the proposed model does not. Such subjective and questionable aspects have been partly responsible for the credibility of these credit agencies being diminished, especially after the 2007-2008 crisis.

show abstract

“…In this context, an important line of research related to clustering stems from the fact that, in some problems, the clusters intrinsically overlap with each other and, consequently, conventional crisp clustering algorithms are not suitable for dealing with this overlap [ 9 , 10 ]. In these cases when an object can “partially” belong to different groups, fuzzy clustering algorithms have been proposed as a powerful methodology in recent years, more flexible than traditional crisp approaches and with excellent results in different real problems [ 11 , 12 ].…”

Section: Introductionmentioning

confidence: 99%

An Island Grouping Genetic Algorithm for Fuzzy Partitioning Problems

Salcedo‐Sanz

Ser²,

Geem

2014

The Scientific World Journal

View full text Add to dashboard Cite

This paper presents a novel fuzzy clustering technique based on grouping genetic algorithms (GGAs), which are a class of evolutionary algorithms especially modified to tackle grouping problems. Our approach hinges on a GGA devised for fuzzy clustering by means of a novel encoding of individuals (containing elements and clusters sections), a new fitness function (a superior modification of the Davies Bouldin index), specially tailored crossover and mutation operators, and the use of a scheme based on a local search and a parallelization process, inspired from an island-based model of evolution. The overall performance of our approach has been assessed over a number of synthetic and real fuzzy clustering problems with different objective functions and distance measures, from which it is concluded that the proposed approach shows excellent performance in all cases.

show abstract

A Corporate Credit Rating Model Using Support Vector Domain Combined with Fuzzy Clustering Algorithm

Cited by 11 publications

References 32 publications

Interpretable Multiclass Models for Corporate Credit Rating Capable of Expressing Doubt

Interpretable Multiclass Models for Corporate Credit Rating Capable of Expressing Doubt

Sovereign Credit Risk Assessment with Multiple Criteria Using an Outranking Method

An Island Grouping Genetic Algorithm for Fuzzy Partitioning Problems

Contact Info

Product

Resources

About