When facing certain problems in science, engineering or technology, it is not enough to find a solution, but it is essential to seek and find the best possible solution through optimization. In many cases the exact optimization procedures are not applicable due to the great computational complexity of the problems. As an alternative to exact optimization, there are approximate optimization algorithms, whose purpose is to reduce computational complexity by pruning some areas of the problem search space. To achieve this, researchers have been inspired by nature, because animals and plants tend to optimize many of their life processes. The purpose of this research is to design a novel bioinspired algorithm for numeric optimization: the Mexican Axolotl Optimization algorithm. The effectiveness of our proposal was compared against nine optimization algorithms (artificial bee colony, cuckoo search, dragonfly algorithm, differential evolution, firefly algorithm, fitness dependent optimizer, whale optimization algorithm, monarch butterfly optimization, and slime mould algorithm) when applied over four sets of benchmark functions (unimodal, multimodal, composite and competition functions). The statistical analysis shows the ability of Mexican Axolotl Optimization algorithm of obtained very good optimization results in all experiments, except for composite functions, where the Mexican Axolotl Optimization algorithm exhibits an average performance.

The Lernmatrix is a classic associative memory model. The Lernmatrix is capable of executing the pattern classification task, but its performance is not competitive when compared to state-of-the-art classifiers. The main contribution of this paper consists of the proposal of a simple mathematical transform, whose application eliminates the subtractive alterations between patterns. As a consequence, the Lernmatrix performance is significantly improved. To perform the experiments, we selected 20 datasets that are challenging for any classifier, as they exhibit class imbalance. The effectiveness of our proposal was compared against seven supervised classifiers of the most important approaches (Bayes, nearest neighbors, decision trees, logistic function, support vector machines, and neural networks). By choosing balanced accuracy as a performance measure, our proposal obtained the best results in 10 datasets. The elimination of subtractive alterations makes the new model competitive against the best classifiers, and sometimes beats them. After applying the Friedman test and the Holm post hoc test, we can conclude that within a 95% confidence, our proposal competes successfully with the most effective classifiers of the state of the art.

An accurate development effort prediction (measured in person-hours) of a software project remains a challenge in the software engineering management field, due to either under-prediction or over-prediction negatively affecting the budgeting and resource allocation to be assigned to the project. Hundreds of studies have explored software development effort prediction models; however, only a limited number of them have applied metaheuristic techniques to optimize the parameters of statistical regression equations (SREs). In addition, most algorithms require a huge amount of data to successfully learn the underlying model parameters. Accordingly, we introduce a novel variant of a swarm intelligence algorithm termed Mexican Axolotl Optimization (MAO), designed for optimizing the SRE parameters and for selecting the SRE within a unified optimization framework. Our approach enables the simultaneous identification of the most suitable SRE model and its optimal configuration, thereby enhancing predictive accuracy. Our proposal was evaluated on eight data sets of software projects selected from an international public repository. The performance of our proposal, named MAO-SRE, is compared to that obtained from traditional SRE, multilayer perceptron neural networks, support vector regression, random forest, and SREs optimized through particle swarm optimization and genetic algorithms. Results based on statistical significance allow us to conclude that MAO-SRE can be used to predict the development effort of software projects, considering their development type, development platform, and programming language generation.

Computational intelligence models applied to medicine have become a growing area of research around the world. This article presents the results of a documentary research that allows identifying the most important computational intelligence models of the state of the art that are applied in the pre-diagnosis of diseases, as well as a study of the algorithms that represent each of these models. Also, through the use of the WEKA platform and the KEEL and UCI repositories, the performance exhibited by these pattern classifiers when applied in the pre-diagnosis of some important chronic diseases in the context of human health is studied. In particular, algorithms representing the most appreciated approaches in the area of intelligent pattern classification are tested. In the experimental part of this article, these computational intelligence algorithms were applied in databases of breast cancer, hyperthyroidism, and hypothyroidism. The results that were derived from the experiments show the superiority of the performance of the vector support machines, the 1-NN classifier and the Lernmatrix tau[9]. The performances achieved allow confirming, with a high degree of certainty, the usefulness of computational intelligence models in the pre-diagnosis of chronic diseases.