We consider the problem of combining ranking results from various sources. In the context of the Web, the main applications include building meta-search engines, combining ranking functions, selecting documents based on multiple criteria, and improving search precision through word associations. Wedevelop a set of techniques for the rank aggregation problem and compare their performance to that of well-known methods. A primary goal of our work is to design rank aggregation techniques that can effectively combat "spam," a serious problem in Web searches. Experiments show that our methods are simple, efficient, and effective. Keywords: rank aggregation, ranking functions, metasearch, multi-word queries, spam 1.

Abstract. Watching television tends to be a social activity. So, adaptive television needs to adapt to groups of users rather than to individual users. In this paper, we discuss different strategies for combining individual user models to adapt to groups, some of which are inspired by Social Choice Theory. In a first experiment, we explore how humans select a sequence of items for a group to watch, based on data about the individuals’ preferences. The results show that humans use some of the strategies such as the Average Strategy (a.k.a. Additive Utilitarian), the Average Without Misery Strategy and the Least Misery Strategy, and care about fairness and avoiding individual misery. In a second experiment, we investigate how satisfied people believe they would be with sequences chosen by different strategies, and how their satisfaction corresponds with that predicted by a number of satisfaction functions. The results show that subjects use normalization, deduct misery, and use the ratings in a non-linear way. One of the satisfaction functions produced reasonable, though not completely correct predictions. According to our subjects, the sequences produced by five strategies give satisfaction to all individuals in the group. The results also show that subjects put more emphasis than expected on showing the best rated item to each individual (at a cost of misery for another individual), and that the ratings of the first and last items in the sequence are especially important. In a final experiment, we explore the influence viewing an item can have on the ratings of other items. This is important for deciding the order in which to present items. The results show an effect of both mood and topical relatedness.

Ranking is an important property that needs to be fully supported by current relational query engines. Recently, several rank-join query operators have been proposed based on rank aggregation algorithms. Rank-join operators progressively rank the join results while performing the join operation. The new operators have a direct impact on traditional query processing and optimization. We introduce a rank-aware query optimization framework that fully integrates rank-join operators into relational query engines. The framework is based on extending the System R dynamic programming algorithm in both enumeration and pruning. We define ranking as an interesting property that triggers the generation of rank-aware query plans. Unlike traditional join operators, optimizing for rank-join operators depends on estimating the input cardinality of these operators. We introduce a probabilistic model for estimating the input cardinality, and hence the cost of a rank-join operator. To our knowledge, this paper is the first effort in estimating the needed input size for optimal rank aggregation algorithms. Costing ranking plans, although challenging, is key to the full integration of rank-join operators in real-world query processing engines. We experimentally evaluate our framework by modifying the query optimizer of an open-source database management system. The experiments show the validity of our framework and the accuracy of the proposed estimation model. 1.

We study the problem of aggregating partial rankings. This problem is motivated by applications such as meta-searching and information retrieval, search engine spam fighting, e-commerce, learning from experts, analysis of population preference sampling, committee decision making and more. We improve recent constant factor approximation algorithms for aggregation of full rankings and generalize them to partial rankings. Our algorithms improved constant factor approximation with respect to all metrics discussed in Fagin et al’s recent important work on comparing partial rankings. We pay special attention to two important types of partial rankings: the well-known top-m lists and the more general p-ratings which we define. We provide first evidence for hardness of aggregating them for constant m, p.

Given a set of rankings, the task of ranking fusion is the problem of combining these lists in such a way to optimize the performance of the combination. The ranking fusion problem is encountered in many situations and, e.g., metasearch is a prominent one. It deals with the problem of combining the result lists returned by multiple search engines in response to a given query, where each item in a result list is ordered with respect to a search engine and a relevance score. Several ranking fusion methods have been proposed in the literature. They can be classified based on whether: (i) they rely on the rank; (ii) they rely on the score; and (iii) they require training data or not. Our paper will make the following contributions: (i) we will report experimental results for the Markov chain rank based methods, for which no large experimental tests have yet been made; (ii) while it is believed that the rank based method, named Borda Count, is competitive with score based methods, we will show that this is not true for metasearch; and (iii) we will show that Markov chain based methods compete with score based methods. This is especially important in the context of metasearch as scores are usually not available from the search engines.

To enable information integration, schema matching is a critical step for discovering semantic correspondences of attributes across heterogeneous sources. While complex matchings are common, because of their far more complex search space, most existing techniques focus on simple 1:1 matchings. To tackle this challenge, this article takes a conceptually novel approach by viewing schema matching as correlation mining, for our task of matching Web query interfaces to integrate the myriad databases on the Internet. On this “deep Web, ” query interfaces generally form complex matchings between attribute groups (e.g., {author} corresponds to {first name, last name} in the Books domain). We observe that the co-occurrences patterns across query interfaces often reveal such complex semantic relationships: grouping attributes (e.g., {first name, last name}) tend to be co-present in query interfaces and thus positively correlated. In contrast, synonym attributes are negatively correlated because they rarely co-occur. This insight enables us to discover complex matchings by a correlation mining approach. In particular, we develop the DCM framework, which consists of data preprocessing, dual mining of positive and negative correlations, and finally matching construction. We evaluate the DCM framework on manually extracted interfaces and the results show good accuracy for discovering complex matchings. Further, to automate the

In pattern recognition, there is a growing use of multiple classifier combinations with the goal to increase recognition performance. In many cases, plurality voting is a part of the combination process. In this article, we discuss and test several well known voting methods from politics and economics on classifier combination in order to see if an alternative to the simple plurality vote exists. We found that, assuming a number of prerequisites, better methods are available, that are comparatively simple and fast. 1

The central problem for many applications in Information Retrieval is ranking and learning to rank is considered as a promising approach for addressing the issue. Ranking SVM, for example, is a state-of-the-art method for learning to rank and has been empirically demonstrated to be effective. In this paper, we study the issue of learning to rank, particularly the approach of using SVM techniques to perform the task. We point out that although Ranking SVM is advantageous, it still has shortcomings. Ranking SVM employs a single hyperplane in the feature space as the model for ranking, which is too simple to tackle complex ranking problems. Furthermore, the training of Ranking SVM is also computationally costly. In this paper, we look at an alternative approach to Ranking SVM, which we call “Multiple Hyperplane Ranker ” (MHR), and make comparisons between the two approaches. MHR takes the divide-and-conquer strategy. It employs multiple hyperplanes to rank instances and finally aggregates the ranking results given by the hyperplanes. MHR contains Ranking SVM as a special case, and MHR can overcome the shortcomings which Ranking SVM suffers from. Experimental results on two information retrieval datasets show that MHR can outperform Ranking SVM in ranking.

The rank aggregation problem is to combine many different rank orderings on the same set of candidates, or alternatives, in order to obtain a "better" ordering. Rank aggregation has been studied extensively in the context of social choice theory, where several "voting paradoxes" have been discovered.

Abstract—Classifier combination holds the potential of improving performance by combining the results of multiple classifers. For domains with very large numbers of classes, such as biometrics, we present an axiomatic framework of desirable mathematical properties for combination functions of rank-based classifiers. This framework represents a continuum of combination rules, including the Borda Count, Logistic Regression, and Highest Rank combination methods as extreme cases [11], [23], [4], [13]. Intuitively, this framework captures how the two complementary concepts of general preference for specific classifiers and the confidence it has in any specific result (as indicated by ranks) can be balanced while maintaining consistent rank interpretation. Mixed Group Ranks (MGR) is a new combination function that balances preference and confidence by generalizing these other functions. We demonstrate that MGR is an effective combination approach by performing multiple experiments on data sets with large numbers of classes and classifiers from the FERET face recognition study. Index Terms—Classification, classifier combination, ensemble methods, sensor fusion, biometrics, face recognition, mixed group ranks, logistic regression, Borda count, highest rank, voting methods. 1