Researchers are defining new types of interactions between humans and machine learning algorithms generically called human-in-the-loop machine learning. Depending on who is in control of the learning process, we can identify: active learning, in which the system remains in control; interactive machine learning, in which there is a closer interaction between users and learning systems; and machine teaching, where human domain experts have control over the learning process. Aside from control, humans can also be involved in the learning process in other ways. In curriculum learning human domain experts try to impose some structure on the examples presented to improve the learning; in explainable AI the focus is on the ability of the model to explain to humans why a given solution was chosen. This collaboration between AI models and humans should not be limited only to the learning process; if we go further, we can see other terms that arise such as Usable and Useful AI. In this paper we review the state of the art of the techniques involved in the new forms of relationship between humans and ML algorithms. Our contribution is not merely listing the different approaches, but to provide definitions clarifying confusing, varied and sometimes contradictory terms; to elucidate and determine the boundaries between the different methods; and to correlate all the techniques searching for the connections and influences between them.

The main objective of this work is to obtain a method that achieves the best accuracy results with a low false positive rate in the classification of K-complexes, a kind of transient waveform found in the Electroencephalogram. With this in mind, the capabilities of several machine learning techniques were tried. The inputs for the models were a set of features based on amplitude and duration measurements obtained from waveforms to be classified. Among all the classifiers tested, the Support Vector Machine obtained the best results with an accuracy of 88.69%. Finally, to enhance the generalization capabilities of the classifiers, while at the same time discarding the existing irrelevant features, feature selection methods were employed. After this process, the classification performance was significantly improved. The best result was obtained applying a correlation-based filter, achieving a 91.40% of accuracy using only 36% of the total input features. (C) 2015 Elsevier Inc. All rights reserved.