Data mining (DM) (Han, Kamber, & Pei, 2011; Tan, Steinbach, & Kumar, 2006) can be described as the most important stage of the knowledge discovery of the data (KDD) process. It consists of the automatic process to discover interesting and unknown trends, patterns, and relationships on datasets, which otherwise would remain undetected. In other words, it tries to reveal the hidden information underlying large amounts of data. The wide range of DM techniques typically involve learning methods from the areas of machine learning (ML), statistics, and database systems, which depend on the type of DM problem being solved. For example, classification by neural networks (NNs) is usually solved by gradient descent network training, whereas decision trees are usually constructed by an iterative process that divides the data into subsets based on conditions that are set on the values of the problem attributes. Even though evolutionary algorithms (EAs; Eiben & Smith, 2003) are not learning techniques, they have been also applied for learning and knowledge extraction. EAs are widely used optimization techniques that allow solving almost any combinatorial or continuous optimization problem, and even those involving both. These search techniques, which are population-based algorithms inspired on natural evolution and genetic processes, can be used as a complement to the standard DM learning approaches or even to replace them, since they can evolve descriptive or predictive models to their optimal structure or parameters. This application of EAs to the DM process is nowadays an important part of what is widely known as evolutionary data mining (EDM). Techniques that have been termed EAs have increased over time (Brabazon, O’Neill, & McGarraghy, 2015). During the 1960s and 1970s, evolution strategies (ESs), genetic algorithms (GAs), evolutionary programming (EP), and genetic programming (GP) were considered as the initial EAs. Lately in the 1980s and 1990s, learning classifier systems (LCSs) and differential evolution (DE) were also included as part of the EAs, considering this group of techniques as the origin of the so named evolutionary computation (EC) (Eiben & Smith, 2003) or evolutionary computing. Since they have been the most used in the literature, and therefore the most widely applied to DM, we will mainly focus on this set of techniques, belonging to the branch of EAs, in this contribution. However, nowadays there are many other types of evolutionary-inspired algorithms, that even though they do not fit with the EA’s previous definition since they are not inspired on natural evolution and genetic processes, they are still based on populations or sets of solutions that cooperatively evolve toward a final optimum implementing intelligent behaviors, social interactions, etc. These more recent algorithms (with respect to the initial EAs) are nowadays considered together with the EAs as the EC (Yang, 2014) current family of algorithms. Since they truly represent a potential improvement to the EDM area, we will also pay attention to the “emergent” application on DM of these evolutionary techniques (Xing & Gao, 2014), namely emergent-EC algorithms from now on, by also analyzing their recent impact with respect to the application of the historically more used EAs.