As the pace of urbanization is accelerating, increasing amount of floodplain has been projected as the future cities. Subsequently, urban flooding is being studied by global emergency management exports due to its inc...As the pace of urbanization is accelerating, increasing amount of floodplain has been projected as the future cities. Subsequently, urban flooding is being studied by global emergency management exports due to its increasingly significant impact on us. Some existing research on flooding emergency management based on the case-based reasoning(CBR) method have made tremendous progress, but the urban flooding case with its stratified data characteristics is required a new methodology which is different from the ones applied to flash floods. So, based on the case-based reasoning(CBR) method, this paper proposed a CPIE-CBR model with four layers, classification filtration, punctiform similarity, interval similarity and entropy weight method, to calculate the case similarity among the urban flooding case with stratified data characteristics. Then we carry out the numerical simulation with the real data about China and conduct some comparison with original ways so that we observe the validity and efficiency of our model in the end.展开更多
Credit risk assessment is an important task of risk management for financial institutions.Machine learning-based approaches have made promising progress in credit risk assessment by treating it as imbalanced binary cl...Credit risk assessment is an important task of risk management for financial institutions.Machine learning-based approaches have made promising progress in credit risk assessment by treating it as imbalanced binary classification tasks.However,few efforts have been made to deal with the class overlap problem that accompanies imbalances simultaneously.To this end,this study proposes a Tomek link and genetic algorithm(GA)-based under-sampling framework(TEUS)to address the class imbalance and overlap issues in binary credit classification by eliminating majority class instances with considering multi-perspective factors.TEUS first determines boundary majority instances with Tomek link,then take the distance from each majority instance to its nearest boundary as the radius and assigns the density of opposite class samples within the radius as the overlap potential of that majority instance.Second,TEUS weighs each non-borderline majority instance based on its information contribution in estimating class labels.After partitioning non-borderline majority instances into subgroups according to overlap potential and information contribution,TEUS applies GA to select samples from subgroups and merge them with the minority samples into a new training set.Innovatively,the design of the fitness function in GA and the grouping of the non-borderline majority not only trade off the multi-perspective characteristics of instances but also help reduce the computational complexity of the sampling optimization search.Numerical experiments on real-world credit data sets demonstrate the effectiveness of the proposed TEUS.展开更多
Discretionary services typically refer to professional work and complex service work by physicians,software developers,web designers,lawyers,or financial analysts,where there are no standard working processes and cust...Discretionary services typically refer to professional work and complex service work by physicians,software developers,web designers,lawyers,or financial analysts,where there are no standard working processes and customer perceived quality of service increases with the time spent on it.Recently,research on these services,especially the corresponding speed-quality tradeoff problem,has gained more and more attention.This paper reviews both the analytical models and the empirical studies in this area,highlighting their contributions and pointing out potential directions for future research.展开更多
基金Supported by Beijing Natural Science Foundation(9162003)
文摘As the pace of urbanization is accelerating, increasing amount of floodplain has been projected as the future cities. Subsequently, urban flooding is being studied by global emergency management exports due to its increasingly significant impact on us. Some existing research on flooding emergency management based on the case-based reasoning(CBR) method have made tremendous progress, but the urban flooding case with its stratified data characteristics is required a new methodology which is different from the ones applied to flash floods. So, based on the case-based reasoning(CBR) method, this paper proposed a CPIE-CBR model with four layers, classification filtration, punctiform similarity, interval similarity and entropy weight method, to calculate the case similarity among the urban flooding case with stratified data characteristics. Then we carry out the numerical simulation with the real data about China and conduct some comparison with original ways so that we observe the validity and efficiency of our model in the end.
基金supported by National Key R&D Programof ChinaunderGrant No.2019YFB1404600Beijing Natural Science Funds under Grant No.9162003Beijing's"High-grade,Precision and Advanced Discipline Construction(Municipal)-Business Administration"project under Grant No.19008022065.
文摘Credit risk assessment is an important task of risk management for financial institutions.Machine learning-based approaches have made promising progress in credit risk assessment by treating it as imbalanced binary classification tasks.However,few efforts have been made to deal with the class overlap problem that accompanies imbalances simultaneously.To this end,this study proposes a Tomek link and genetic algorithm(GA)-based under-sampling framework(TEUS)to address the class imbalance and overlap issues in binary credit classification by eliminating majority class instances with considering multi-perspective factors.TEUS first determines boundary majority instances with Tomek link,then take the distance from each majority instance to its nearest boundary as the radius and assigns the density of opposite class samples within the radius as the overlap potential of that majority instance.Second,TEUS weighs each non-borderline majority instance based on its information contribution in estimating class labels.After partitioning non-borderline majority instances into subgroups according to overlap potential and information contribution,TEUS applies GA to select samples from subgroups and merge them with the minority samples into a new training set.Innovatively,the design of the fitness function in GA and the grouping of the non-borderline majority not only trade off the multi-perspective characteristics of instances but also help reduce the computational complexity of the sampling optimization search.Numerical experiments on real-world credit data sets demonstrate the effectiveness of the proposed TEUS.
基金supported by the National Natural Science Foundation of China under grant number 72071204the Open Research Fund of Collaborative Innovation Centre for State-owned Assets Administration,Beijing Technology and Business University under grant number BTBUGZGL201903Beijing social science funds under grant number 17GLB011.
文摘Discretionary services typically refer to professional work and complex service work by physicians,software developers,web designers,lawyers,or financial analysts,where there are no standard working processes and customer perceived quality of service increases with the time spent on it.Recently,research on these services,especially the corresponding speed-quality tradeoff problem,has gained more and more attention.This paper reviews both the analytical models and the empirical studies in this area,highlighting their contributions and pointing out potential directions for future research.