Abstract
We present a study in the area of sentiment analysis of clients' commentaries transcribed by assistants of a help-desk service of one Portuguese telecommunications company. We have adopted a lexicon-based approach to determine the polarity of the sentiment of each commentary, based on the so called opinion words. This task was by no means easy, as not many tools are available for the Portuguese language. The initial results with the off-the-shelf solutions were rather poor. This has motivated us to carry out a number of enhancements, including, for instance, enriching the given lexicon with domain specific terms, formulating specific rules for negation and amplifiers. Automatic pruning of some of the lexicon terms has led to a significant improvement in performance. As our final system achieved a very good performance, our work should be of interest to others working on domain specific solutions for languages where ready-made solutions are not available.

Abstract

Abstract

Abstract

Abstract
The combination of information obtained from data mining technique from physicochemical and organoleptic data analysis allowed similarities between the wines of the nine sub-regions in the Demarcated Region of Vinho Verde. Through clustering techniques, four clusters were identified, each characterized by its centroid. The measure of information gain, together with supervised rule-based learning, was used to find the differentiating characteristics. This study allowed the interconnection of the characteristics of the wines of these sub-regions, which can improve the decision making on the profiles of these same wines.

Abstract
Selecting appropriate classification algorithms for a given dataset is crucial and useful in practice but is also full of challenges. In order to maximize performance, users of machine learning algorithms need methods that can help them identify the most relevant features in datasets, select algorithms and determine their appropriate hyperparameter settings. In this paper, a method of recommending classification algorithms is proposed. It is oriented towards the average ranking method, combining algorithm rankings observed on prior datasets to identify the best algorithms for a new dataset. Our method uses a special case of data mining workflow that combines algorithm selection preceded by a feature selection method (CFS).