基于非平衡数据集的支持向量域分类模型，提出了一种银行客户个人信用预测方法。
A new predication method of customer credit of Banks is proposed based on the support vector domain classification model of non-balance data set.
针对此问题，本研究对结合支持向量机（SVM）算法的几种常用非平衡数据分类方法进行实验比较，包括投票整合分类器和移动分类面等。
Combined with support vector machine (SVM) algorithm, several common approaches to deal with the unbalanced problem were compared, including majority voting, the moving boundary surface, etc.
运用我国2001~2005民营上市公司的非平衡面板数据，考察了民营金字塔结构下终极所有者掏空底层公司的影响因素。
The paper analyzes the influential factors to tunneling of ultimate owner in private pyramid firms using the unbalanced panel data of Chinese private listed firms from 2001 to 2005.
对非平衡传输中采集一级数据进行了定量的时间和容量分析，得出了具有应用价值的结论。
Also, the time and capacity of collecting first-level data in an unbalanced transmission procedure are analyzed quantitatively, with a valuable conclusion on the application of this protocol formed.
有缺失数据两因素随机区组试验资料是两因素水平组合重复数不等的非平衡资料。
The data with missing values in two-factor experiment with randomized block design is the unbalanced data with unequal replications of level combination of two factors.
在采样数据反馈控制中，只要利用一个参数便将系统控制到非稳定平衡点。
The single parameter sampled data feedback control directs the system to unstable equilibrium point with only one parameter.
针对体系的非理想过程，采用UNIFAC基团贡献法计算液液平衡常数，解决了实验数据不足造成的局限。
As to the non-linear process of the system, we calculate the constant of the Liquid-liquid equilibrium by making use of the UNIFAC Method to solve the problem of the lack of experimental data.
鉴于基金流量对基金管理者和股市监管者的重要性，本文利用非平衡面板数据实证分析了基金家族特征是否影响基金流量。
Given the importance of fund flows for fund managers and the supervisors of stock market, this paper examines whether fund families affect fund flows based on unbalanced panel data.
系统主要由宽带光源、光纤光栅、非平衡马赫-曾德干涉仪、数据采集与处理等部分组成。
The system mainly consists of wide-band light source, fiber Bragg grating, unbalanced Mach-Zehnder interferometer, and data acquisition and processing components.
该算法可以关联非共轭的二元体系液液平衡数据，并且对参数初值的选取要求较低。
The new algorithm could be associated with the non-line liquid-liquid equilibrium data, and the requirement of initial parameters was lower.
确定了低温下非平衡载流子复合率及扩散速度的降低，是导致SRAM断电后数据残留的主要原因。
The root cause of data remanence of SRAM is determined which is the decrease of excess-carrier recombination rate and carrier diffusion velocity at low temperature.
摘要：最近在机器学习和数据挖掘上，非平衡类问题成为了一个研究热点。
Abstract: Recently, the problem of Class-imbalance has become a hotspot in machine learning and data mining.
根据所测非平衡张力，对应动态表面压力数据也进行了计算。
The dynamic surface pressure was obtained according to experimental data.
然后介绍了非平衡类数据下分类算法性能的评价措施。
Section 4 introduces several performance measures for evaluating classifier algorithms in the presence of imbalanced data sets.
然后介绍了非平衡类数据下分类算法性能的评价措施。
Section 4 introduces several performance measures for evaluating classifier algorithms in the presence of imbalanced data sets.
应用推荐