Abstract
Replicated time series are a particular type of repeated measures, which consist of time-sequences of measurements taken from several subjects (experimental units). We consider independent replications of count time series that are modelled by first-order integer-valued autoregressive processes, INAR(1). In this work, we propose several estimation methods using the classical and the Bayesian approaches and both in time and frequency domains. Furthermore, we study the asymptotic properties of the estimators. The methods are illustrated and their performance is compared in a simulation study. Finally, the methods are applied to a set of observations concerning sunspot data.

Abstract
Spatial and temporal sea level variability in the North Atlantic is investigated from Topex/Poseidon (T/P) altimetry data. Time series of sea level anomalies on a regular 5 degrees grid are analysed. Non-linear denoising through thresholding in the wavelet transform domain is carried out for each series in order to remove noise while preserving non-smooth features. Principal Component Analysis (PCA) is used to obtain a spatio-temporal description of the sea level field, To avoid modal mixing and improve interpretation of the principal modes, PCA is implemented separately for seasonal and trend components of the sea level field obtained from a wavelet-based multiresolution analysis. The leading pattern of the seasonal field reflects the dominance of a stable annual cycle over the study area and the change in the seasonal regime approaching the equator with contribution of the semi-annual cycle and phase-shift in the annual cycle in the tropical Atlantic. The leading pattern of the trend field is a broad spatial pattern associated with North Atlantic Oscillation (NAO), reflecting the influence of atmospheric conditions on interannual sea level variability.

Abstract
In this paper we propose a method for grouping and summarizing large sets of association rules according to the items contained in each rule. We use hierarchical clustering to partition the initial rule set into thematically coherent subsets. This enables the summarization of the rule set by adequately choosing a representative rule for each subset, and helps in the interactive exploration of the rule model by the user. We define the requirements of our approach, and formally show the adequacy of the chosen approach to our aims. Rule clusters can also be used to infer novel interest measures for the rules. Such measures are based on the lexicon of the rules and are complementary to measures based on statistical properties, such as confidence, lift and conviction. We show examples of the application of the proposed techniques.

Abstract
This Doctoral Consortium paper focuses on Extreme Adaptivity, a set of top level requirements for adaptive hypertext systems, which has resulted from one year of examining the adaptive hypertext landscape. The complete specification of a system, KnowledgeAtoms, is also given, mainly as an example of Extreme Adaptivity. Additional methodological elements are discussed.

Abstract
In this paper we describe an application of recommender systems to team building in a company or organization. The recommender system uses a collaborative filtering model based approach. Recommender models are sets of association rules extracted from the activity log of employees assigned to projects or tasks. Recommendation is performed at two levels: first by recommending a single team element given a partially built team; and second by recommending changes to a completed team. The methodology is applied to a case study with real data. The results are evaluated through experimental tests and one survey to potential users.

Abstract
The Support Vector Machine algorithm is sensitive to the choice of parameter settings. If these are not set correctly, the algorithm may have a substandard performance. Suggesting a good setting is thus an important problem. We propose a meta-learning methodology for this purpose and exploit information about the past performance of different settings. The methodology is applied to set the width of the Gaussian kernel. We carry out an extensive empirical evaluation, including comparisons with other methods (fixed default ranking; selection based on cross-validation and a heuristic method commonly used to set the width of the SVM kernel). We show that our methodology can select settings with low error while providing significant savings in time. Further work should be carried out to see how the methodology could be adapted to different parameter setting tasks.