Treffer: Data classification using reversible decimal first-degree cellular automata.

This work explores how the cycles in the configuration space of a reversible decimal first-degree cellular automaton can be used as a tool for classifying any dimensional dataset. When selecting a reversible cellular automata for data classification, it has to satisfy different properties such as a low rate of information propagation and properties related to cycle structure. In this study, we identify reversible cellular automata that are suitable for this purpose. To convert the dataset instances into a format suitable for the configuration of decimal cellular automata, we implement two methods: bin encoding and concatenation with splitting. According to the majority training label for the configurations present in a cycle, all configurations within the cycle are assigned to a specific class. We validate the performance of our model by conducting various experiments using 12 benchmark datasets and comparing the results with those of other machine learning models. We show that the majority of datasets achieve an accuracy above 90%. It is also observed that our reversible decimal cellular automaton-based classifier, which uses the Bin-encoding method, performs at par with all existing machine learning models. [ABSTRACT FROM AUTHOR]

Copyright of Natural Computing is the property of Springer Nature and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)