Choosing the most suitable algorithm to perform a machine learning task for a new problem is a recurrent and complex task. In multi-target regression tasks, when problem transformation methods are applied, this choice is even harder. The reason is the need to simultaneously choose the problem transformation method and the base learning algorithm. This work investigates how to bridge the gap of method/base learner recommendation for problems with multiple outputs. In meta-learning experiments, we use a large number of multi-target regression datasets to investigate whether using meta-learning can provide good recommendations. To do this, we compared the meta-models induced by 3 different ML algorithms, including three variations for each of them, and selected 58 meta-features that we believe are relevant for extracting good dataset descriptions for the meta-learning process. In the experimental results, the meta-models outperformed the baselines (Majority and Random) by recommending the most suitable solution for multi-target regression (for the transformation method and base-learner) with high predictive performance, including real-world applications. The meta-features and the relation between the transformation method and base-learner provided important insights regarding the optimal problem transformation method. Furthermore, when comparing the application of algorithm adaptation and problem transformation methods, our meta-learning proposal was capable of statistically overcoming all competitors, which resulted in a predictive performance using the best choice per problem.

Imbalanced datasets may negatively impact the predictive performance of most classical classification algorithms. This problem, commonly found in real-world, is known in machine learning domain as imbalanced learning. Most techniques proposed to deal with imbalanced learning have been proposed and applied only to binary classification. When applied to multiclass tasks, their efficiency usually decreases and negative side effects may appear. This paper addresses these limitations by presenting a novel adaptive approach, E-MOSAIC (Ensemble of Classifiers based on MultiObjective Genetic Sampling for Imbalanced Classification). E-MOSAIC evolves a selection of samples extracted from training dataset, which are treated as individuals of a MOEA. The multiobjective process looks for the best combinations of instances capable of producing classifiers with high predictive accuracy in all classes. E-MOSAIC also incorporates two mechanisms to promote the diversity of these classifiers, which are combined into an ensemble specifically designed for imbalanced learning. Experiments using twenty imbalanced multi-class datasets were carried out. In these experiments, the predictive performance of E-MOSAIC is compared with state-of-the-art methods, including methods based on presampling, active-learning, cost-sensitive, and boosting. According to the experimental results, the proposed method obtained the best predictive performance for the multiclass accuracy measures mAUC and G-mean.

Community detection looks for groups of nodes in networks, mainly using network topological, link-based features, not taking into account features associated with each node. Clustering algorithms, on the other hand, look for groups of objects using features describing each object. Recently, link features and node attributes have been combined to improve community detection. Community detection methods can be designed to identify communities that are disjoint or overlapping, crisp or soft and static or dynamic. In this paper, we propose a dynamic community detection method for finding soft overlapping groups in temporal networks with node attributes. Our approach is based on a non-negative matrix factorization model that uses automatic relevance determination to detect the number of communities. Preliminary results on toy and artificial networks, are promising. To the extent of our knowledge, a dynamic approach that includes link and node information, for soft overlapping community detection, has not been proposed before.

Meta-learning has been successfully used for algorithm recommendation tasks. It uses machine learning to induce meta-models able to predict the best algorithms for a new dataset. In this paper, meta-models are applied to a set of meta-features, describing a dataset, to predict the performance of clustering algorithms applied to this dataset. The paper also proposes a new set of meta-features, based on correlation and dissimilarity measures. Experimental results show that these meta-features improve the recommendation. Additionally, this paper evaluates the importance of each meta-feature for the recommendation.

As Collaborative Filtering becomes increasingly important in both academia and industry recommendation solutions, it also becomes imperative to study the algorithm selection task in this domain. This problem aims at finding automatic solutions which enable the selection of the best algorithms for a new problem, without performing full-fledged training and validation procedures. Existing work in this area includes several approaches using Metalearning, which relate the characteristics of the problem domain with the performance of the algorithms. This study explores an alternative approach to deal with this problem. Since, in essence, the algorithm selection problem is a recommendation problem, we investigate the use of Collaborative Filtering algorithms to select Collaborative Filtering algorithms. The proposed approach integrates subsampling landmarkers, a data characterization approach commonly used in Metalearning, with a Collaborative Filtering methodology, named CF4CF. The predictive performance obtained by CF4CF using benchmark recommendation datasets was similar or superior to that obtained with Metalearning.

The problem of information overload motivated the appearance of Recommender Systems.From the several open problems in this area, the decision of which is the best recommendation algorithm for a specific problem is one of the most important and less studied.The current trend to solve this problem is the experimental evaluation of several recommendation algorithms in a handful of datasets.However, these studies require an extensive amount of computational resources, particularly processing time.To avoid these drawbacks, researchers have investigated the use of Metalearning to select the best recommendation algorithms in different scopes.Such studies allow to understand the relationships between data characteristics and the relative performance of recommendation algorithms, which can be used to select the best algorithm(s) for a new problem.The contributions of this study are two-fold: 1) to identify and discuss the key concepts of algorithm selection for recommendation algorithms via a systematic literature review and 2) to perform an experimental study on the Metalearning approaches reviewed in order to identify the most promising concepts for automatic selection of recommendation algorithms.

This work addresses the problem of selecting Tensor Factorization algorithms for the Context-aware Filtering recommendation task using a metalearning approach. The most important challenge of applying metalearning on new problems is the development of useful measures able to characterize the data, i.e. metafeatures. We propose an extensive and exhaustive set of metafeatures to characterize Context-aware Filtering recommendation task. These metafeatures take advantage of the tensor's hierarchical structure via slice operations. The algorithm selection task is addressed as a Label Ranking problem, which ranks the Tensor Factorization algorithms according to their expected performance, rather than simply selecting the algorithm that is expected to obtain the best performance. A comprehensive experimental work is conducted on both levels, baselevel and metalevel (Tensor Factorization and Label Ranking, respectively). The results show that the proposed metafeatures lead to metamodels that tend to rank Tensor Factorization algorithms accurately and that the selected algorithms present high recommendation performance.

Data stream mining is an emergent research area that investigates knowledge extraction from large amounts of continuously generated data, produced by non-stationary distribution. Novelty detection, the ability to identify new or previously unknown situations, is a useful ability for learning systems, especially when dealing with data streams, where concepts may appear, disappear, or evolve overtime. There are several studies currently investigating the application of novelty detection techniques in data streams. However, there is no consensus regarding how to evaluate the performance of these techniques. In this study, we propose a new evaluation methodology for multiclass novelty detection in data streams able to deal with: i) unsupervised learning, which generates novelty patterns without an association with the true classes, where one class may be composed of a novelty set, ii) confusion matrix that increases overtime, iii) confusion matrix with a column representing unknown examples, i.e., those not explained by the model, and iv) representation of the evaluation measures overtime. We propose a new methodology to associate the novelty patterns detected by the algorithm, in an unsupervised fashion, with the true classes. Finally, we evaluate the performance of the proposed methodology through the use of known novelty detection algorithms with artificial and real data sets.