This paper describes a model for optimum information retrieval over a distributed document collection. The model stems from Robertson's Probability Ranking Principle: Having computed individual document rankings correlated to diierent subcollections, these local rankings are stepwise merged into a nal ranking list where the documents are ordered according to their probability of relevance. Here...

We study the standard retrieval task of ranking a fixed set of documents given a previously unseen query and pose it as the half-transductive ranking problem. The task is partly transductive as the document set is fixed. Existing transductive approaches are natural non-linear methods for this set, but have no direct out-ofsample extension. Functional approaches, on the other hand, can be applie...

The field of information retrieval deals with finding relevant documents from a large document collection or the World Wide Web in response to a user’s query seeking relevant information. Ranking functions play a very important role in the retrieval performance of such retrieval systems and search engines. A single ranking function does not perform well across different user queries, and docume...

In this paper we examine the problem of ranking candidate expansion terms for query expansion. We show, by an extension to the traditional F4 scheme, how partial relevance assessments (how relevant a document is) and ostensive evidence (when a document was assessed relevant) can be incorporated into a term ranking function. We then investigate this new term ranking function in three user experi...

We address the problem of learning to rank documents in a multilingual context, when reference ranking information is only partially available. We propose a multiview learning approach to this semisupervised ranking task, where the translation of a document in a given language is considered as a view of the document. Although both multiview and semi-supervised learning of classifiers have been ...

We propose an efficient text summarization technique that involves two basic operations. The first operation involves finding coherent chunks in the document and the second operation involves ranking the text in the individual coherent chunks and picking the sentences that rank above a given threshold. The coherent chunks are formed by exploiting the lexical relationship between adjacent senten...

In the context of learning to rank for information retrieval [15], we study a general class of “DCG-like” ranking loss functions which include DCG [13] and approximate ERR [6] as specific cases. We then study the Bayes optimal ranking function for this class, which is a function of the conditional distribution of graded document relevance levels. Our main contribution is a novel class of rankin...

We study the standard retrieval task of ranking a fixed set of documents given a previously unseen query and pose it as the half-transductive ranking problem. The task is partly transductive as the document set is fixed. Existing transductive approaches are natural non-linear methods for this set, but have no direct out-ofsample extension. Functional approaches, on the other hand, can be applie...

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...

The rapid growth of the Web has led to the development of many techniques for enhancing search rankings by using precomputed numeric document attributes such as the estimated popularity or importance of Web pages. For efficient keyword-search query processing over large document repositories, it is vital that these auxiliary attribute vectors, containing numeric per-document properties, be kept...