This paper concerns several important topics of the Symmetry journal, namely, pattern recognition, computer-aided design, diversity and similarity. We also take advantage of the symmetric structure of a membership function. Searching for the (sub) optimal subset of features is an NP-hard problem. In this paper, a binary swallow swarm optimization (BSSO) algorithm for feature selection is proposed. To solve the classification problem, we use a fuzzy rule-based classifier. To evaluate the feature selection performance of our method, BSSO is compared to induction without feature selection and some similar algorithms on well-known benchmark datasets. Experimental results show the promising behavior of the proposed method in the optimal selection of features.

In this paper, we propose six methods for binarization of the swallow swarm optimization (SSO) algorithm to solve the feature selection problem. The relevance of the selected feature subsets is estimated by two classifiers: a fuzzy rule-based classifier and a classifier based on k-nearest neighbors. To find an optimal subset of features, we take into account the number of features and classification accuracy. The developed algorithms are tested on datasets from the KEEL repository. For the statistical evaluation of the binarization methods, we use Friedman’s two-way analysis of variance by ranks for related samples. The best feature selection result is shown by a hybrid method based on modified algebraic operations and MERGE operation introduced by the authors of this paper. The best classification accuracy is achieved with a V-shaped transfer function.

In this paper we propose modifications of the well-known algorithm of particle swarm optimization (PSO). These changes affect the mapping of the motion of particles from continuous space to binary space for searching in it, which is widely used to solve the problem of feature selection. The modified binary PSO variations were tested on the dataset SVC2004 dedicated to the problem of user authentication based on dynamic features of a handwritten signature. In the example of k-nearest neighbours (kNN), experiments were carried out to find the optimal subset of features. The search for the subset was considered as a multicriteria optimization problem, taking into account the accuracy of the model and the number of features.

Biometric systems have numerous advantages over traditional authentication methods and should be taken into account: uniqueness, fraud-resistance, portability, convenience of use, scalability, possibility of reliable attendance recording. Use of electroencephalography (EEG) data as biometric factor can keep all these benefits and add new ones. This paper provides insight how EEG based biometrics identification ability can be affected by the number of participants, their gender and age. In order to study aforementioned factor's influence Leipzig Study for Mind-Body-Emotion Interactions (LEMON) resting state (both eyes-open state and eyes-closed state) EEG dataset were used. After few preprocessing adjustments and applying discrete wavelet transform as time-frequency domain feature extraction method resulting dataset were used for classification task. The testing was conducted in 5-fold cross-validation mode using the shallow model, Linear Support Vector Classifier. From experiment results following conclusions were made: increased number of participants decreased classifier accuracy, with age increase accuracy steadily improved, gender effects on accuracy are inconclusive.