卷积重提取特征的文档列表排序学习方法  

Listwise Reranking via Convolutional Re-extracted Features

作　　者：曹军梅[1] 马乐荣[1] CAO Junmei;MA Lerong(College of Mathematics and Computer Science,Yan′an University,Yan’an,Shaanxi 716000,China)

机构地区：[1]延安大学数学与计算机科学学院,陕西延安716000

出　　处：《中文信息学报》2020年第8期86-93,共8页Journal of Chinese Information Processing

基　　金：国家自然科学基金(61751217,61866038);延安大学科研引导项目(YDY2018-11)。

摘　　要：在许多信息检索任务中,为了进一步提高检索性能,通常需要对检索到的文档进行重新排序,目前的排序学习方法主要集中在损失函数的构造上,而没有考虑特征之间的关系。该文将多通道深度卷积神经网络作用于文档列表排序学习方法,即ListCNN,实现了信息检索的精确重排序。由于从文档中提取的多个特征中有一些特征具有局部相关性和冗余性,因此,文中使用卷积神经网络来重新提取特征,以提高列表方法的性能。ListCNN架构考虑了原始文档特征的局部相关性,能够有效地重新提取代表性特征。在公共数据集LETOR 4.0上对ListCNN进行实验验证,结果表明其性能优于已有文档列表方法。Re-ranking retrieved documents are usually required to further improve the performance in many information retrieval tasks.In this paper,we conduct multi-channel deep convolutional neural networks(CNNs)on listwise approaches for learning to rank,namely ListCNN.For the multi-modal features extracted from documents,we find that some features are locally correlated with redundancy.Accordingly,we propose to employ deep neural networks(i.e.,modified CNNs)to re-extract features to boost the performance of classical listwise approaches.Validated on public datasets of LETOR 4.0,the proposed ListCNN architecture demonstrates superior performance for re-ranking in comparison with other state-of-the-arts methods.

关 键 词：排序学习 文档列表 梯度下降 卷积神经网络 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]