Abstract
Online social networks present themselves as valuable information sources about their users and their respective behaviours and interests. Many researchers in data mining have analysed these types of data, aiming to find interesting patterns. This paper addresses the problem of identifying and displaying tweet profiles by analysing multiple types of data: spatial, temporal, social and content. The data mining process that extracts the patterns is composed by the manipulation of the dissimilarity matrices for each type of data, which are fed to a clustering algorithm to obtain the desired patterns. This paper studies appropriate distance functions for the different types of data, the normalization and combination methods available for different dimensions and the existing clustering algorithms. The visualization platform is designed for a dynamic and intuitive usage, aimed at revealing the extracted profiles in an understandable and interactive manner. In order to accomplish this, various visualization patterns were studied and widgets were chosen to better represent the information. The use of the project is illustrated with data from the Portuguese twittosphere.

Abstract
This paper proposes a framework to decompose and develop metafeatures for Metalearning (MtL) problems. Several metafeatures (also known as data characteristics) are proposed in the literature for a wide range of problems. Since MtL applicability is very general but problem dependent, researchers focus on generating specific and yet informative metafeatures for each problem. This process is carried without any sort of conceptual framework. We believe that such framework would open new horizons on the development of metafeatures and also aid the process of understanding the metafeatures already proposed in the state-of-the-art. We propose a framework with the aim of fill that gap and we show its applicability in a scenario of algorithm recommendation for regression problems.

Abstract

Abstract
The huge amount of online information deprives the user to keep up with his/hers interests and preferences, Recommender Systems appeared to solve this problem, by employing social behavioural paradigms in order to recommend potentially interesting items to users, Among the several kinds of Recommender Systems, one of the most mature and most used in real world applications are known as Collaborative Filtering. These methods recommend items based on the preferences of similar-users, using only a user-item rating matrix. In this pa™ per we explain a methodology to use Multi™Agent based simulation to study the evolution of the data rating matrix and its effect on the performance of several Collaborative Filtering algorithms. Our results show that the best performing methods are user-based and item-based Collaborative Filtering and that the average algorithm performance is surprisingly constant for different rating schemes.

Abstract
Dynamic real-world applications that generate data continuously have introduced new challenges for the machine learning community, since the concepts to be learned are likely to change over time. In such scenarios, an appropriate model at a time point may rapidly become obsolete, requiring updating or replacement. As there are several learning algorithms available, choosing one whose bias suits the current data best is not a trivial task. In this paper, we present a meta-learning based method for periodic algorithm selection in time-changing environments, named MetaStream. It works by mapping the characteristics extracted from the past and incoming data to the performance of regression models in order to choose between single learning algorithms or their combination. Experimental results for two real regression problems showed that MetaStream is able to improve the general performance of the learning system compared to a baseline method and an ensemble-based approach.

Abstract
Support Vector Machines (SVMs) have achieved a considerable attention due to their theoretical foundations and good empirical performance when compared to other learning algorithms in different applications. However, the SVM performance strongly depends on the adequate calibration of its parameters. In this work we proposed a hybrid multi-objective architecture which combines meta-learning (ML) with multi-objective particle swarm optimization algorithms for the SVM parameter selection problem. Given an input problem, the proposed architecture uses a ML technique to suggest an initial Pareto front of SVM configurations based on previous similar learning problems; the suggested Pareto front is then refined by a multi-objective optimization algorithm. In this combination, solutions provided by ML are possibly located in good regions in the search space. Hence, using a reduced number of successful candidates, the search process would converge faster and be less expensive. In the performed experiments, the proposed solution was compared to traditional multi-objective algorithms with random initialization, obtaining Pareto fronts with higher quality on a set of 100 classification problems.