Pattern recognition (PR) is the study of how a system can observe the environment, learn to distinguish patterns of interest from their background, and make decisions about their classification or categorization. In general, a pattern can be any entity described with features, where the dimensionality of the feature space can range from being a few to thousands. The four best approaches for PR are template matching, statistical classification, syntactic or structural recognition, and artificial neural networks (ANNs). The latter approach attempts to use some organizational principles such as learning, generalization, adaptivity, fault tolerance, distributed representation, and computation in order to achieve the recognition. The main 746characteristics of ANNs are that they have the ability to learn complex nonlinear input-output relationships, use sequential training procedures, and adapt themselves to data. Some popular models of ANNs have been shown to be capable of associative memory (AM) and learning. The learning process involves updating the network architecture and modifying the weights between the neurons so that the network can efficiently perform a specific classification/clustering task.
School of Computer Science

Oommen, J, Qin, K. (Ke), & Calitoiu, D. (Dragos). (2017). The science and art of chaotic pattern recognition. In Handbook of Applications of Chaos Theory (pp. 745–802). doi:10.1201/b20232