Detecting non-human activity in social networks has become an area of great interest for both industry and academia. In this context, obtaining a high detection accuracy is not the only desired quality; experts in the application domain would also like having an understandable model, with which one may explain a decision. An explanatory decision model may help experts to consider, for example, taking legal action against an account that has displayed offensive behavior, or forewarning an account holder about suspicious activity. In this paper, we shall use a pattern-based classification mechanism to social bot detection, specifically for Twitter. Furthermore, we shall introduce a new feature model for social bot detection, which extends (part of) an existing model with features out of Twitter account usage and tweet content sentiment analysis. From our experimental results, we shall see that our mechanism outperforms other, state-of-the-art classifiers, not based on patterns; and that our feature model yields better classification results than others reported on in the literature.

Decision trees are popular as stand-alone classifiers or as base learners in ensemble classifiers. Mostly, this is due to decision trees having the advantage of being easy to explain. To improve the classification performance of decision trees, some authors have used Multivariate Decision Trees (MDTs), which allow combinations of features when splitting a node. While there is growing interest in the area, recent research in MDTs all have in common that they do not provide adequate comparison of related work: they do not consider relevant rival techniques, or they test algorithm performance in an insufficient number of databases. As a result, claims have no statistical sustain and, hence, there is a lack of general understanding of the actual capabilities of existing MDT induction algorithms, crucial to improving the state-of-the-art. In this paper, we report on an exhaustive review of MDTs. In particular, we give an overview of 37 MDT induction algorithms, out of which we have experimentally compared 19 of them in 57 databases. We provide a statistical comparison in all databases and subsets of databases according to the number of classes, number of features, number of instances, and degree of class imbalance. This allows us to identify groups of top-performing algorithms for different types of databases.