面向排序学习的特征分析的研究

排序是信息检索中一个重要的环节,当今已经提出百余种用于构建排序函数的特征,如何利用这些特征构建更有效的排序函数成为当今的一个热点问题,因此排序学习（Learning to Rank）,一个信息检索与机器学习的交叉学科,越来越受到人们的重视。从排序特征的构建方式易知,特征之间并不是完全独立的,然而现有的排序学习方法的研究,很少在特征分析的基础上,从特征重组与选择的角度,来构建更有效的排序函数。针对这一问题,提出如下的模型框架：对构建排序函数的特征集合进行分析,然后重组与选择,利用排序学习方法学习排序函数。基于这一框架,提出四种特征处理的算法：基于主成分分析的特征重组方法、基于MAP、前向选择和排序学习算法隐含的特征选择。实验结果显示,经过特征处理后,利用排序学习算法构建的排序函数,一般优于原始的排序函数。     

A machine learning approach to query generation in plagiarism source retrieval

Plagiarism source retrieval is the core task of plagiarism detection. It has become the standard for plagiarism detection to use the queries extracted from suspicious documents to retrieve the plagiarism sources. Generating queries from a suspicious document is one of the most important steps in plagiarism source retrieval. Heuristic-based query generation methods are widely used in the current research. Each heuristic-based method has its own advantages, and no one statistically outperforms the others on all suspicious document segments when generating queries for source retrieval. Further improvements on heuristic methods for source retrieval rely mainly on the experience of experts. This leads to difficulties in putting forward new heuristic methods that can overcome the shortcomings of the existing ones. This paper paves the way for a new statistical machine learning approach to select the best queries from the candidates. The statistical machine learning approach to query generation for source retrieval is formulated as a ranking framework. Specifically, it aims to achieve the optimal source retrieval performance for each suspicious document segment. The proposed method exploits learning to rank to generate queries from the candidates. To our knowledge, our work is the first research to apply machine learning methods to resolve the problem of query generation for source retrieval. To solve the essential problem of an absence of training data for learning to rank, the building of training samples for source retrieval is also conducted. We rigorously evaluate various aspects of the proposed method on the publicly available PAN source retrieval corpus. With respect to the established baselines, the experimental results show that applying our proposed query generation method based on machine learning yields statistically significant improvements over baselines in source retrieval effectiveness.     

Boosted ranking models: a unifying framework for ranking predictions

Ranking is an important functionality in a diverse array of applications, including web search, similarity-based multimedia retrieval, nearest neighbor classification, and recommendation systems. In this paper, we propose a new method, called Boosted Ranking Model (BRM), for learning how to rank from training data. An important feature of the proposed method is that it is domain-independent and can thus be applied to a wide range of ranking domains. The main contribution of the new method is that it reduces the problem of learning how to rank to the much more simple, and well-studied problem of constructing an optimized binary classifier from simple, weak classifiers. Using that reduction, our method constructs an optimized ranking model using multiple simple, easy-to-define ranking models as building blocks. The new method is a unifying framework that includes, as special cases, specific methods that we have proposed in earlier publications for specific ranking applications, such as nearest neighbor retrieval and classification. In this paper, we reformulate those earlier methods as special cases of the proposed BRM method, and we also illustrate a novel application of BRM, on the problem of making movie recommendations to individual users.     

面向排序学习的锦标赛排序特征选择方法

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一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。     

基于查询性能预测的鲁棒检索排序研究

信息检索技术致力于从海量的信息资源中为用户获取所需的信息。相较于传统的简单模型,近些年来的大量研究工作在提升了检索结果平均质量的同时,往往忽略了鲁棒性的问题,即造成了很多查询的性能下降,导致用户满意度的显著下降。本文提出了一种基于排序学习的查询性能预测方法,针对每一个查询,对多种模型得到的检索结果列表进行预测,将其中预测性能最优的检索结果列表展示给用户。在LETOR的三个标准数据集OHSUMED、MQ2008和MSLR-WEB10K上的一系列对比实验表明,在以经典的BM25模型作为基准的情况下,与当前最好的检索模型之一LambdaMART相比,该方法在提升了检索结果平均质量的同时,显著地减少了性能下降的查询的数量,具备较好的鲁棒性。
     

Training query filtering for semi-supervised learning to rank with pseudo labels

Semi-supervised learning is a machine learning paradigm that can be applied to create pseudo labels from unlabeled data for learning a ranking model, when there is only limited or no training examples available. However, the effectiveness of semi-supervised learning in information retrieval (IR) can be hindered by the low quality pseudo labels, hence the need for the training query filtering that removes the low quality queries. In this paper, we assume two application scenarios with respect to the availability of human labels. First, for applications without any labeled data available, a clustering-based approach is proposed to select the high quality training queries. This approach selects the training queries following the empirical observation that the relevant documents of high quality training queries are highly coherent. Second, for applications with limited labeled data available, a classification-based approach is proposed. This approach learns a weak classifier to predict the retrieval performance gain of a given training query by making use of query features. The queries with high performance gains are selected for the following transduction process to create the pseudo labels for learning to rank algorithms. Experimental results on the standard LETOR dataset show that our proposed approaches outperform the strong baselines.     

面向微博搜索的时间敏感的排序学习方法

近年来微博检索已经成为信息检索领域的研究热点。相关的研究表明,微博检索具有时间敏感性。已有工作根据不同的时间敏感性假设,例如,时间越新文档越相关,或者时间越接近热点时刻文档越相关,得到多种不同的检索模型,都在一定程度上提高了检索效果。但是这些假设主要来自于观察,是一种直观简化的假设,仅能从某个方面反映时间因素影响微博排序的规律。该文验证了微博检索具有复杂的时间敏感特性,直观的简化假设并不能准确地描述这种特性。在此基础上提出了一个利用微博的时间特征和文本特征,通过机器学习的方式来构建一个针对时间敏感的微博检索的排序学习模型(TLTR)。在时间特征上,考察了查询相关的全局时间特征以及查询-文档对的局部时间特征。在TREC Microblog Track 20112012数据集上的实验结果表明,TLTR模型优于现有的其他时间敏感的微博排序方法。     

Semi-supervised ranking for document retrieval

Ranking functions are an important component of information retrieval systems. Recently there has been a surge of research in the field of “learning to rank”, which aims at using labeled training data and machine learning algorithms to construct reliable ranking functions. Machine learning methods such as neural networks, support vector machines, and least squares have been successfully applied to ranking problems, and some are already being deployed in commercial search engines.Despite these successes, most algorithms to date construct ranking functions in a supervised learning setting, which assume that relevance labels are provided by human annotators prior to training the ranking function. Such methods may perform poorly when human relevance judgments are not available for a wide range of queries. In this paper, we examine whether additional unlabeled data, which is easy to obtain, can be used to improve supervised algorithms. In particular, we investigate the transductive setting, where the unlabeled data is equivalent to the test data.We propose a simple yet flexible transductive meta-algorithm: the key idea is to adapt the training procedure to each test list after observing the documents that need to be ranked. We investigate two instantiations of this general framework: The Feature Generation approach is based on discovering more salient features from the unlabeled test data and training a ranker on this test-dependent feature-set. The importance weighting approach is based on ideas in the domain adaptation literature, and works by re-weighting the training data to match the statistics of each test list. We demonstrate that both approaches improve over supervised algorithms on the TREC and OHSUMED tasks from the LETOR dataset.     

Multi‐level reranking approach for bug localization

Bug fixing has a key role in software quality evaluation. Bug fixing starts with the bug localization step, in which developers use textual bug information to find location of source codes which have the bug. Bug localization is a tedious and time consuming process. Information retrieval requires understanding the programme's goal, coding structure, programming logic and the relevant attributes of bug. Information retrieval (IR) based bug localization is a retrieval task, where bug reports and source files represent the queries and documents, respectively. In this paper, we propose BugCatcher, a newly developed bug localization method based on multi‐level re‐ranking IR technique. We evaluate BugCatcher on three open source projects with approximately 3400 bugs. Our experiments show that multi‐level reranking approach to bug localization is promising. Retrieval performance and accuracy of BugCatcher are better than current bug localization tools, and BugCatcher has the best Top N, Mean Average Precision (MAP) and Mean Reciprocal Rank (MRR) values for all datasets.     

Transductive learning to rank using association rules

Learning to rank, a task to learn ranking functions to sort a set of entities using machine learning techniques, has recently attracted much interest in information retrieval and machine learning research. However, most of the existing work conducts a supervised learning fashion. In this paper, we propose a transductive method which extracts paired preference information from the unlabeled test data. Then we design a loss function to incorporate this preference data with the labeled training data, and learn ranking functions by optimizing the loss function via a derived Ranking SVM framework. The experimental results on the LETOR 2.0 benchmark data collections show that our transductive method can significantly outperform the state-of-the-art supervised baseline.     

View-wised discriminative ranking for 3D object retrieval

In this paper, we propose a new framework which can capture the latent relative information within the multiple views of 3D model, named View-wised Discriminative Ranking(VDR). Different to existing view-based methods which treat the multiple views as the independent information, we want to model the relative information within multiple views. By placing the views of model in certain order, we learn the parameters of ranking function as a new robust model representation. We evaluate our proposal on several challenging datasets for 3D retrieval and the comparison experiments demonstrate the superiority of the proposed method in both retrieval accuracy and efficiency.     

A discriminative kernel-based approach to rank images from text queries

This paper introduces a discriminative model for the retrieval of images from text queries. Our approach formalizes the retrieval task as a ranking problem, and introduces a learning procedure optimizing a criterion related to the ranking performance. The proposed model hence addresses the retrieval problem directly and does not rely on an intermediate image annotation task, which contrasts with previous research. Moreover, our learning procedure builds upon recent work on the online learning of kernel-based classifiers. This yields an efficient, scalable algorithm, which can benefit from recent kernels developed for image comparison. The experiments performed over stock photography data show the advantage of our discriminative ranking approach over state-of-the-art alternatives (e.g. our model yields 26.3% average precision over the Corel dataset, which should be compared to 22.0%, for the best alternative model evaluated). Further analysis of the results shows that our model is especially advantageous over difficult queries such as queries with few relevant pictures or multiple-word queries.     

基于查询聚类的排序学习算法

排序学习算法作为信息检索与机器学习的一个交叉领域,越来越受到人们的重视。然而,几乎没有排序学习算法考虑到查询差异的存在。文中查询被建模为多元高斯分布,KL距离被用来度量查询之间的距离,利用谱聚类方法对查询进行聚类,为每个聚类类别训练一个排序函数。实验结果表明经过聚类得到的排序函数需要较少的训练样例,但是它的性能却和没有经过聚类得到的排序函数具有可比性,甚至优于后者。     

基于查询语义特性的稠密文本检索模型

针对现有的稠密文本检索模型（dense passage retrieval,DPR）存在的负采样效率低、易产生过拟合等问题,提出了一种基于查询语义特性的稠密文本检索模型（Q-DPR）。首先,针对模型的负采样过程,提出了一种基于近邻查询的负采样方法。该方法通过检索近邻查询,快速地构建高质量的负相关样本,以降低模型的训练成本。其次,针对模型易产生过拟合的问题,提出了一种基于对比学习的查询自监督方法。该方法通过建立查询间的自监督对比损失,缓解模型对训练标签的过拟合,从而提升模型的检索准确性。Q-DPR在面向开放领域问答的大型数据集MSMARCO上表现优异,取得了0.348的平均倒数排名以及0.975的召回率。实验结果证明,该模型成功地降低了训练的开销,同时也提升了检索的性能。     

An adaptive learning automata-based ranking function discovery algorithm

Due to the massive amount of heterogeneous information on the web, insufficient and vague user queries, and use of the same query by different users for different aims, the information retrieval process deals with a huge amount of uncertainty and doubt. Under such circumstances, designing an efficient retrieval function and ranking algorithm by which the most relevant results are provided is of the greatest importance. In this paper, a learning automata-based ranking function discovery algorithm in which different sources of information are combined is proposed. In this method, the learning automaton is used to adjust the portion of the final ranking that is assigned to each source of evidence based on the user feedback. All sources of information are first given the same importance. The proportion of a given source increases, if the documents provided by this source are reviewed by the user and decreases otherwise. As the proposed algorithm proceeds, the probability of appearance of each source in the final ranking gets proportional to its relevance to the user queries. Several simulation experiments are conducted on well-known data collections and query types to show the performance of the proposed algorithm. The obtained results demonstrate that the proposed algorithm outperforms several existing methods in terms of precision at position n, mean average precision, and normalized discount cumulative gain.     

Latent ranking analysis using pairwise comparisons in crowdsourcing platforms

Ranking items is an essential problem in recommendation systems. Since comparing two items is the simplest type of queries in order to measure the relevance of items, the problem of aggregating pairwise comparisons to obtain a global ranking has been widely studied. Furthermore, ranking with pairwise comparisons has recently received a lot of attention in crowdsourcing systems where binary comparative queries can be used effectively to make assessments faster for precise rankings. In order to learn a ranking based on a training set of queries and their labels obtained from annotators, machine learning algorithms are generally used to find the appropriate ranking model which describes the data set the best.In this paper, we propose a probabilistic model for learning multiple latent rankings by using pairwise comparisons. Our novel model can capture multiple hidden rankings underlying the pairwise comparisons. Based on the model, we develop an efficient inference algorithm to learn multiple latent rankings as well as an effective inference algorithm for active learning to update the model parameters in crowdsourcing systems whenever new pairwise comparisons are supplied. The performance study with synthetic and real-life data sets confirms the effectiveness of our model and inference algorithms.     

基于强化学习的多样性文档排序算法

在排序学习方法中,通过直接优化信息检索评价指标来学习排序模型的方法,取得了很好的排序效果,但是其损失函数在利用所有排序位置信息以及融合多样性排序因素方面还有待提高。为此,提出基于强化学习的多样性文档排序算法。首先,将强化学习思想应用于文档排序问题,通过将排序行为建模为马尔可夫决策过程,在每一次迭代过程中利用所有排序位置的信息,不断为每个排序位置选择最优的文档。其次,在排序过程中结合多样性策略,依据相似度阈值,裁剪高度相似的文档,从而保证排序结果的多样性。最后,在公共数据集上的实验结果表明,提出的算法在保证排序准确性的同时,增强了排序结果的多样性。     

Rank-Order Correlation-Based Feature Vector Context Transformation for Learning to Rank for Information Retrieval

As a crucial task in information retrieval, ranking defines the preferential order among the retrieved documents for a given query. Supervised learning has recently been dedicated to automatically learning ranking models by incorporating various models into one effective model. This paper proposes a novel supervised learning method, in which instances are represented as bags of contexts of features, instead of bags of features. The method applies rank-order correlations to measure the correlation relationships between features. The feature vectors of instances, i.e., the 1st-order raw feature vectors, are then mapped into the feature correlation space via projection to derive the context-level feature vectors, i.e., the 2nd-order context feature vectors. As for ranking model learning, Ranking SVM is employed with the 2nd-order context feature vectors as the input. The proposed method is evaluated using the LETOR benchmark datasets and is found to perform well with competitive results. The results suggest that the learning method benefits from the rank-order-correlation-based feature vector context transformation.     

一种基于排序学习方法的查询扩展技术

查询扩展作为一门重要的信息检索技术,是以用户查询为基础,通过一定策略在原始查询中加入一些相关的扩展词,从而使得查询能够更加准确地描述用户信息需求。排序学习方法利用机器学习的知识构造排序模型对数据进行排序,是当前机器学习与信息检索交叉领域的研究热点。该文尝试利用伪相关反馈技术,在查询扩展中引入排序学习算法,从文档集合中提取与扩展词相关的特征,训练针对于扩展词的排序模型,并利用排序模型对新查询的扩展词集合进行重新排序,将排序后的扩展词根据排序得分赋予相应的权重,加入到原始查询中进行二次检索,从而提高信息检索的准确率。在TREC数据集合上的实验结果表明,引入排序学习算法有助于提高伪相关反馈的检索性能。     

Continuous ranking on uncertain streams

Data uncertainty widely exists in many web applications, financial applications and sensor networks. Ranking queries that return a number of tuples with maximal ranking scores are important in the field of database management. Most existing work focuses on proposing static solutions for various ranking semantics over uncertain data. Our focus is to handle continuous ranking queries on uncertain data streams: testing each new tuple to output highly-ranked tuples. The main challenge comes from not only the fact that the possible world space will grow exponentially when new tuples arrive, but also the requirement for low space- and time-complexity to adapt to the streaming environments. This paper aims at handling continuous ranking queries on uncertain data streams. We first study how to handle this issue exactly, then we propose a novel method (exponential sampling) to estimate the expected rank of a tuple with high quality. Analysis in theory and detailed experimental reports evaluate the proposed methods.