当样本量较大时，QIF方法表明工作相关阵的选择等价于无偏估计方程的选择。
When the sample size is great, QIF approach shows that the selection of working correlation is same to the selection of unbiased estimating equation.
在相对合理的样本空间中采用偏最小二乘回归建立模型。
Then it builds a model by partial least squares regression on the obtained sample space.
该算法将主动式学习、有偏分类和增量学习结合起来，对相关反馈过程中的小样本有偏学习问题进行建模。
The algorithm combines active learning, biased classification and incremental learning to model the small sample biased learning problem in relevance feedback process.
应用推荐