排序问题在信息检索领域是一个非常重要的课题。虽然排序学习模型的算法早已被深入研究，但针对排序学习算法中的特征选择的研究却很少。现实的情况是，许多用于分类的特征选择方法被直接应用到排序学习中。但由于排序和分类有着显著的差异，应研究出针对排序的特征选择算法。文中在介绍常用的排序学习的特征选择方法的基础上，提出了一种全新的、适用于QA问题的排序学习的特征选择方法-锦标赛排序特征选择方法。实验结果显示，这种新的特征选择方法在提高特征提取效率和降低特征向量维数方面都有显著改善。%Ranking is a very important topic in information retrieval. And algorithms for learning to ranking models have been intensively studied,this is not the case for feature selection,despite of its importance. The reality is that many feature selection methods used in classi-fication are directly applied to ranking. Argue that because of the striking differences between ranking and classification,it is better to de-velop different feature selection methods for ranking. To this end,a new feature selection method in this paper is proposed. The feature se-lection based on common feature selection methods,a novel method adapting better to learning to rank of QA problems,the champion-ships sort feature selection method is proposed in this paper. The experimental results show that,this method can improve the efficiency of feature selection and reduce the dimensions of the feature vector.
展开▼
