The number of activities provided by scientific workflow management systems is large, which requires scientists to know many of them to take advantage of the reusability of these systems. To minimize this problem, the literature presents some techniques to recommend activities during the scientific workflow construction. In this paper we specified and developed a hybrid activity recommendation system considering information on frequency, input and outputs of activities and ontological annotations. Additionally, this paper presents a modeling of activities recommendation as a classification problem, tested using 5 classifiers; 5 regressors; and a composite approach which uses a Support Vector Machine (SVM) classifier, combining the results of other classifiers and regressors to recommend; and Rotation Forest, an ensemble of classifiers. The proposed technique was compared to related techniques and to classifiers and regressors, using 10-fold-cross-validation, achieving a Mean Reciprocal Rank (MRR) at least 70% greater than those obtained by classical techniques.