Support Vector Machines with Evolutionary Feature Selection for Default Prediction

Härdle, Wolfgang Karl; Prastyo, Dedy Dwi; Hafner, Christian

doi:10.2139/ssrn.2894201

Cited by 5 publications

(4 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this method, to properly assess the decision function, the separator function or indeed the regression function, the data which has labels or goals is used to learn. Forward to training with sufficient data, machine learning, instance SVM, could be used to predict or identify what decisions have been made if new data is entered and also the result is unspecified [10]. Unless the performance either at time of testing by itself doesn't perform as expected, the parameters of the machine learning function can be modified to boost the accuracy of the model.…”

Section: Introductionmentioning

confidence: 99%

Using Hierarchical Likelihood Towards Support Vector Machine: Theory and Its Application

et al. 2020

View full text Add to dashboard Cite

The H-likelihood method proposed by Lee and Nelder (1996) is extensively used in a wide range of data. In terms of direction, repetitive measured data within classification can be examined employing hierarchical generalized linear models (HGLMs). Whether we are concerned in multiple endpoints which are correlated, instead Multivariate Double Hierarchical Generalized Linear Models (DHGLM) can be taken into consideration. This paper addresses the implementation of this principle to vector selection and support machines. Based on the analysis with the fish morphology class Sardinella lemuru (Bali sardinella) and setting the best epsilon 0.7 cost 4 parameter reaching best performance: 0.2327401. Predictive value of fish sex was calculated 0.997319 and Region under the curve: 0.8967. At the same time, we extend the large-scale case studies for stress testing of the SVM method by using three datasets from UCI machine learning repository including the Bank marketing dataset, the car evaluation database and Human Activity Recognition Using Smartphones dataset. In a nutshell by employing SVM-DHGLM increased the accuracy, precision, recall, for feature selection and classification. Long story short, the H-likelihood provides an excellent and usable structure for statistical inference of the unobservable general deterministic model, while preserving the advantages of the original probability structure for fixed parameters. We presume that more new groups of models will be created and that the H-likelihood will be commonly used for their inferences and the application in big data and machine learning.

show abstract

Section: Introductionmentioning

confidence: 99%

Using Hierarchical Likelihood Towards Support Vector Machine: Theory and Its Application

et al. 2020

View full text Add to dashboard Cite

show abstract

“…The results indicated that using the rough set in initial variable selection increased prediction accuracy Angelini et alYeh et al employed a rough set and the RF to create a credit rating model featuring high prediction accuracy, and then added DEA to compare the operating efficiency among variables used in forecasting. The results indicated that after DEA was added, the forecasting accuracy was 88%; when only financial data were added, the forecasting accuracy was 76%, revealing that twostep forecasting exhibits superior performanceHärdle et al (2012) SVM BPN Rough set Fuzzy theory Härdle et al combined rough set and fuzzy theory to perform variable selection, and then added the SVM to determine the optimal hyperplane before create credit rating classification. The results indicated that the rough set yielded the highest forecasting rate, 89.42%, followed by SVM, which yielded an 87.72% forecasting rate, and then BPN, which yielded an 81.Cheng integrated the KMV model and DD with financial variables to analyze credit rating by using the RF.…”

mentioning

confidence: 99%

Evaluating credit rating prediction by using the KMV model and random forest

2016

View full text Add to dashboard Cite

Purpose An increasing number of investors have begun using financial data to develop optimal investment portfolios; therefore, the public financial data shared in the capital market plays a critical role in credit ratings. These data enable investors to understand the credit levels of debtors from a bank perspective; this facilitates predicting the debtor default rate to efficiently evaluate investment risks. The paper aims to discuss these issues. Design/methodology/approach A credit rating model can be developed to reduce the risk of adverse selection and moral hazard caused by information asymmetry in the loan market. In this study, a random forest (RF) was used to evaluate financial variables and construct credit rating prediction models. Data-mining techniques, including an RF, decision tree, neural networks, and support vector machine, were used to search for suitable credit rating forecasting methods. The distance to default from the KMV model was then incorporated into the credit rating model as a research variable to increase predictive power of various data-mining techniques. In addition, four-level and nine-level classification were set to investigate the accuracy rates of various models. Findings The experimental results indicated that applying the RF in the variable feature selection process and developing a forecasting model was the most effective method of predicting credit ratings; the four-level and nine-level feature-selection settings achieved 95.5 and 87.8 percent accuracy rates, respectively, indicating that RF demonstrated outstanding feature selection and forecasting capacity. Research limitations/implications The experimental cases were based on financial data from public companies in North America. Practical implications Practical implication of this study indicates the most effective financial variables were dividends common/ordinary, cash dividends, volatility assumption, and risk-free rate assumption. Originality/value The RF model can be used to perform feature selection and efficiently filter numerous financial variables to obtain crediting rating information instantly.

show abstract

“…Previously, the researcher could be using kind of heuristics and metaheuristics optimization [19][20][21][22]. In addition, the split of training and testing data has also influenced the accuracy of the model, such as the percentage 90:10, 80:20, 70:30, 60:40, and 50:50 [23][24][25][26]. On the other hand, there are some techniques to separate the training and testing data using the K fold [27,28].…”

Section: Literature Reviewmentioning

confidence: 99%

Using Machine Learning to Predict Visitors to Totally Protected Areas in Sarawak, Malaysia

et al. 2022

View full text Add to dashboard Cite

The machine learning approach has been widely used in many areas of studies, including the tourism sector. It can offer powerful estimation for prediction. With a growing number of tourism activities, there is a need to predict tourists’ classification for monitoring, decision making, and planning formulation. This paper aims to predict visitors to totally protected areas in Sarawak using machine learning techniques. The prediction model developed would be able to identify significant factors affecting local and foreign visitors to these areas. Several machine learning techniques such as k-NN, Naive Bayes, and Decision Tree were used to predict whether local and foreign visitors’ arrival was high, medium, or low to these totally protected areas in Sarawak, Malaysia. The data of local and foreign visitors’ arrival to eighteen totally protected areas covering national parks, nature reserves, and wildlife centers in Sarawak, Malaysia, from 2015 to 2019 were used in this study. Variables such as the age of the park, distance from the nearest city, types of the park, recreation services availability, natural characteristics availability, and types of connectivity were used in the model. Based on the accuracy measure, precision, and recall, results show Decision Tree (Gain Ratio) exhibited the best prediction performance for both local visitors (accuracy = 80.65) and foreign visitors (accuracy = 84.35%). Distance to the nearest city and size of the park were found to be the most important predictors in predicting the local tourist visitors’ park classification, while for foreign visitors, age, type of park, and the natural characteristics availability were the significant predictors in predicting the foreign tourist visitors’ parks classification. This study exemplifies that machine learning has respectable potential for the prediction of visitors’ data. Future research should consider bagging and boosting algorithms to develop a visitors’ prediction model.

show abstract

Support Vector Machines with Evolutionary Feature Selection for Default Prediction

Cited by 5 publications

References 51 publications

Using Hierarchical Likelihood Towards Support Vector Machine: Theory and Its Application

Using Hierarchical Likelihood Towards Support Vector Machine: Theory and Its Application

Evaluating credit rating prediction by using the KMV model and random forest

Using Machine Learning to Predict Visitors to Totally Protected Areas in Sarawak, Malaysia

Contact Info

Product

Resources

About