本文基于UCI机器学习库中的一个信用卡审批的数据,以是否同意审批为响应变量,以15个描述客户信息的离散和连续指标作为解释变量建立预测模型。提出了一种改进的弹性网损失支持向量机(QCaenSVM)预测模型,旨在提高信用卡审批行业中审批周...本文基于UCI机器学习库中的一个信用卡审批的数据,以是否同意审批为响应变量,以15个描述客户信息的离散和连续指标作为解释变量建立预测模型。提出了一种改进的弹性网损失支持向量机(QCaenSVM)预测模型,旨在提高信用卡审批行业中审批周期长且结果不一致以及数据利用不充分的不足之处。QCaenSVM模型通过融合弹性网损失函数和分位数的概念,优化了传统支持向量机的性能。该模型在含噪声数据环境下具有较好的表现性能,并有效处理了数据中的不确定性。在应用于信用卡的预测实践中,QCaenSVM成功识别出更可能选择审批的客户,明显提高了预测效果,为相关部门和客户群体提供了有力工具。In this paper, we build a predictive model based on data from a credit card approval in the UCI Machine Learning Library, with whether or not to agree to the approval as the response variable, and 15 discrete and continuous metrics describing the customer information as the explanatory variables. An improved Quantile-Capped Asymmetric Elastic Net Support Vector Machine (QCaenSVM) prediction model is proposed to improve the shortcomings of the credit card approval industry in terms of long approval cycles with inconsistent results and underutilization of data. The QCaenSVM model optimizes the performance of the traditional support vector machine by incorporating the concepts of the elastic net loss function and quartiles. The model has better performance in noisy data-containing environments and effectively handles uncertainties in the data. In the prediction practice applied to credit cards, QCaenSVM successfully identifies customers who are more likely to choose approval, significantly improves the prediction effect, and provides a powerful tool for relevant departments and customer groups.展开更多
A new iris feature extraction approach using both spatial and frequency domain is presented. Steerable pyramid is adopted to get the orientation information on iris images. The feature sequence is extracted on each su...A new iris feature extraction approach using both spatial and frequency domain is presented. Steerable pyramid is adopted to get the orientation information on iris images. The feature sequence is extracted on each sub-image and used to train Support Vector Machine (SVM) as iris classifiers. SVM has drawn great interest recently as one of the best classifiers in machine learning, although there is a problem in the use of traditional SVM for iris recognition. It cannot treat False Accept and False Reject differently with different security requirements. Therefore, a new kind of SVM called Non-symmetrical SVM is presented to classify the iris features. Experimental data shows that Non-symmetrical SVM can satisfy various security requirements in iris recognition applications. Feature sequence combined with spatial and frequency domain represents the variation details of the iris patterns properly. The results in this study demonstrate the potential of our new approach, and show that it performs more satis- factorily when compared to former algorithms.展开更多
文摘本文基于UCI机器学习库中的一个信用卡审批的数据,以是否同意审批为响应变量,以15个描述客户信息的离散和连续指标作为解释变量建立预测模型。提出了一种改进的弹性网损失支持向量机(QCaenSVM)预测模型,旨在提高信用卡审批行业中审批周期长且结果不一致以及数据利用不充分的不足之处。QCaenSVM模型通过融合弹性网损失函数和分位数的概念,优化了传统支持向量机的性能。该模型在含噪声数据环境下具有较好的表现性能,并有效处理了数据中的不确定性。在应用于信用卡的预测实践中,QCaenSVM成功识别出更可能选择审批的客户,明显提高了预测效果,为相关部门和客户群体提供了有力工具。In this paper, we build a predictive model based on data from a credit card approval in the UCI Machine Learning Library, with whether or not to agree to the approval as the response variable, and 15 discrete and continuous metrics describing the customer information as the explanatory variables. An improved Quantile-Capped Asymmetric Elastic Net Support Vector Machine (QCaenSVM) prediction model is proposed to improve the shortcomings of the credit card approval industry in terms of long approval cycles with inconsistent results and underutilization of data. The QCaenSVM model optimizes the performance of the traditional support vector machine by incorporating the concepts of the elastic net loss function and quartiles. The model has better performance in noisy data-containing environments and effectively handles uncertainties in the data. In the prediction practice applied to credit cards, QCaenSVM successfully identifies customers who are more likely to choose approval, significantly improves the prediction effect, and provides a powerful tool for relevant departments and customer groups.
基金Project supported by the National Natural Science Foundation of China (No. 60272031), Educational Department Doctor Foundation of China (No. 20010335049), and Zhejiang Provincial Natural ScienceFoundation (No. ZD0212), China
文摘A new iris feature extraction approach using both spatial and frequency domain is presented. Steerable pyramid is adopted to get the orientation information on iris images. The feature sequence is extracted on each sub-image and used to train Support Vector Machine (SVM) as iris classifiers. SVM has drawn great interest recently as one of the best classifiers in machine learning, although there is a problem in the use of traditional SVM for iris recognition. It cannot treat False Accept and False Reject differently with different security requirements. Therefore, a new kind of SVM called Non-symmetrical SVM is presented to classify the iris features. Experimental data shows that Non-symmetrical SVM can satisfy various security requirements in iris recognition applications. Feature sequence combined with spatial and frequency domain represents the variation details of the iris patterns properly. The results in this study demonstrate the potential of our new approach, and show that it performs more satis- factorily when compared to former algorithms.
