We study generalization properties of the area under the ROC curve (AUC), a quantity that has been advocated as an evaluation criterion for the bipartite ranking problem. The AUC is a different term than the error rate used for evaluation in classification problems; consequently, existing generalization bounds for the classification error rate cannot be used to draw conclusions about the AUC. I...

We study generalization properties of the area under an ROC curve (AUC), a quantity that has been advocated as an evaluation criterion for bipartite ranking problems. The AUC is a different and more complex term than the error rate used for evaluation in classification problems; consequently, existing generalization bounds for the classification error rate cannot be used to draw conclusions abo...

This paper summarizes our work on the TREC 2015 Clinical Decision Support Track. We present a customized learningto-rank algorithm and a query term position based re-ranking model to better satisfy the tasks. We design two learning-to-rank framework: the pointwise loss function based on random forest and the pairwise loss function based on SVM. The position based re-ranking model is composed of...

Current learning to rank approaches commonly focus on learning the best possible ranking function given a small fixed set of documents. This document set is often retrieved from the collection using a simple unsupervised bag-of-words method, e.g. BM25. This can potentially lead to learning a sub-optimal ranking, since many relevant documents may be excluded from the initially retrieved set. In ...

The problem of ranking a set of visual samples according to their relevance to a query plays an important role in computer vision. The traditional approach for ranking is to train a binary classifier such as a support vector machine (svm). Binary classifiers suffer from two main deficiencies: (i) they do not optimize a ranking-based loss function, for example, the average precision (ap) loss; a...

One can use the Leiden Rankings for grouping research universities by considering universities which are not significantly different as a homogeneous set. Such groupings reduce the complexity of the rankings without losing information. We pursue this classification using both statistical significance and effect sizes of differences among 902 universities in 54 countries; we focus on the UK, Ger...

Each item in a given collection is characterized by a set of possible performances. A (ranking) method is a function that assigns an ordering of the items to every performance profile. Ranking by Rating consists in evaluating each item’s performance by using an exogenous rating function, and ranking items according to their performance ratings. Any such method is separable: the ordering of two ...

The problem of ranking a set of visual samples according to their relevance to a query plays an important role in computer vision. The traditional approach for ranking is to train a binary classifier such as a support vector machine (svm). Binary classifiers suffer from two main deficiencies: (i) they do not optimize a ranking-based loss function, for example, the average precision (ap) loss; a...

In this paper, we propose a new method for learning to rank. ‘Ranking SVM’ is a method for performing the task. It formulizes the problem as that of binary classification on instance pairs and performs the classification by means of Support Vector Machines (SVM). In Ranking SVM, the losses for incorrect classifications of instance pairs between different rank pairs are defined as the same. We n...

Most protein PageRank studies do not use signal flow direction information in protein interactions because this information was not readily available in large protein databases until recently. Therefore, four questions have yet to be answered: A) What is the general difference between signal emitting and receiving in a protein interactome? B) Which proteins are among the top ranked in direction...