Statistical learning theory was introduced in the late 1960’s. Until the 1990’s it was a purely theoretical analysis of the problem of function estimation from a given collection of data. In the middle of the 1990’s new types of learning algorithms (called support vector machines) based

errorful data. We refer to the original data as RCV1-v1, and the corrected data as RCV1-v2. We benchmark several widely used supervised learning methods on RCV1-v2, illustrating the collection’s properties, suggesting new directions for research, and providing baseline results for future studies. We make

from document collections that requires only a handful of training examples from users. These examples are used to generate extraction patterns, that in turn result in new tuples being extracted from the document collection. We build on this idea and present our Snowball system. Snowball

improve retrieval. We argue that these problems arise only when clustering is used in an attempt to improve conventional search techniques. However, looking at clustering as an information access tool in its own right obviates these objections, and provides a powerful new access paradigm. We present a

amplitudes you told me about, they’re so complicated and absurd, what makes you think those are right? Maybe they aren’t right. ’ Such remarks are obvious and are perfectly clear to anybody who is working on this problem. It does not do any good to point this out.” —Richard Feynman [1, p.161]

for this state of affairs is that existing theories of work The authors express great appreciation to members of the consulting firm that helped us gain access to the organizations where this research was conducted; to Kenneth Brousseau, Daniel Feldman, and Linda Frank for assistance in data collection

The use of information retrieval systems in networked environments raises a new set of issues that have received little attention. These issues include ranking document collections for relevance to a query, selecting the best set of collections from a ranked list, and merging the document rankings

collection’—provide a new resource not only for systematic analyses of unknown gene functions and gene regulatory networks but also for genome-wide testing of mutational effects in a common strain background, E. coli K-12 BW25113. We were unable to disrupt 303 genes, including 37 of unknown function, which

In an earlier paper, we introduced a new “boosting” algorithm called AdaBoost which, theoretically, can be used to significantly reduce the error of any learning algorithm that consistently generates classifiers whose performance is a little better than random guessing. We also introduced

We examine properties of residual-based tests for the null of no cointegration for dynamic panels in which both the short-run dynamics and the long-run slope coefficients are permitted to be heterogeneous across individual members of the panel+ The tests also allow for individual heterogeneous fixed effects and trend terms, and we consider both pooled within dimension tests and group mean between dimension tests+ We derive limiting distributions for these and show that they are normal and free of nuisance parameters+ We also provide Monte Carlo evidence to demonstrate their small sample size and power performance, and we illustrate their use in testing purchasing power parity for the post–Bretton Woods period+ 1.