Abstract
Atmospheric pressure varies within a wide range of scales and thus a multi-scale description of its variability is particularly appealing. In this study, a scale-by-scale analysis of the global sea-level pressure field is carried out from reanalysis data. Wavelet-based analysis of variance is applied in order to describe the variability of the pressure field in terms of patterns representing the contribution of each scale to the overall variance. Signals at the seasonal scales account for the largest fraction of sea-level pressure variance (typically more than 60%) except in the Southern Ocean, in the Equatorial Pacific and in the North Atlantic. In the Southern Ocean and over the North Atlantic, high-frequency signals contribute to a considerable fraction (30-50%) of the overall variance in sea-level pressure. In the Equatorial Pacific, large-scale variability, associated with ENSO, contributes up to 40% of the total variance.

Abstract
Long-term sea-level variability in Chesapeake Bay is examined from long tide gauge records in order to assess the influence of climate factors on sea-level changes in this complex estuarine system. A time series decomposition method based on autoregression is applied to extract flexible seasonal and low-frequency components from the tide gauge records, allowing to analyse long-term sea-level variability not only by estimating linear trends from the records, but also by examining fluctuations in seasonal and long-term patterns. Long-term sea-level variability in Chesapeake Bay shows considerable decadal variability. At the annual scale, variability is mainly determined by atmospheric factors, specifically atmospheric pressure and zonal wind, but no systematic trends are found in the amplitude of the annual cycle. On longer time scales, precipitation rate, a proxy for river discharge, is the main factor influencing decadal sea-level variability. Linear trends in relative sea-level heights range from 2.66 +/- 0.075 mm/year (at Baltimore) to 4.40 +/- 0.086 mm/year (at Hampton Roads) for the 1955-2007 period. Due to the gentle slope of most of the bay margin, a sea-level increase of this magnitude poses a significant threat in terms of wetland loss and consequent environmental impacts.

Abstract
Heart rate variability (HRV) data display nonstationary characteristics, exhibit long-range correlations (memory) and instantaneous variability (volatility). Recently, we have proposed fractionally integrated autoregressive moving average (ARFIMA) models for a parametric alternative to the widely-used technique detrended fluctuation analysis, for long memory estimation in HRV. Usually, the volatility in HRV studies is assessed by recursive least squares. In this work, we propose an alternative approach based on ARFIMA models with generalized autoregressive conditionally heteroscedastic (GARCH) innovations. ARFIMA-GARCH models, combined with selective adaptive segmentation, may be used to capture and remove long-range correlation and estimate the conditional volatility in 24 hour HRV recordings. The ARFIMA-GARCH approach is applied to 24 hour HRV recordings from the Noltisalis database allowing to discriminate between the different groups.

Abstract
In this paper, an evaluation of several methods allowing the estimation of the Harmonic-to-Noise Ratio (HNR) of sustained vowels was conducted. The HNR estimation methods are mainly based on time, spectral, and cepstral signal representations. An algorithm was implemented for each method and was tested with synthesized voice sounds in order to evaluate their accuracy. Tests were also conducted with real pathological voice sounds in order to evaluate the behaviour of the different methods under real conditions. © 2008 Division of Signal Processin.

Abstract
The use of collaborative filtering (CF) recommenders on the Web is typically done in environments where data is constantly flowing. In this paper we propose an incremental version of item-based CF for implicit binary ratings, and compare it with a non-incremental one, as well as with an incremental user-based approach. We also study the usage of sparse matrices in these algorithms. We observe that recall and precision tend to improve when we continuously add information to the recommender model, and that the time spent for recommendation does not degrade. Time for updating the similarity matrix is relatively low and motivates the use of the item-based incremental approach. © 2008 IEEE.

Abstract
Traditionally, recommender systems for the Web deal with applications that have two types of entities/dimensions, users and items. With these dimensions, a recommendation model can be built and used to identify a set of N items that will be of interest to a certain user. In this paper we propose a direct method that enriches the information in the access logs with new dimensions. We empirically test this method with two recommender systems, an item-based collaborative filtering technique and association rules, on three data sets. Our results show that while collaborative filtering is not able to take advantage of the new dimensions added, association rules are capable of profiting from our direct method. © 2008 IEEE.