The set of heuristics constituting the methods of deep learning has proved very efficient in complex problems of artificial intelligence such as pattern recognition, speech recognition, etc., solving them with better accuracy than previously applied methods. Our aim in this work has been to integrate the concept of the rough set to the repository of tools applied in deep learning in the form of rough mereological granular computing. In our previous research we have presented the high efficiency of our decision system approximation techniques (creating granular reflections of systems), which, with a large reduction in the size of the training systems, maintained the internal knowledge of the original data. The current research has led us to the question whether granular reflections of decision systems can be effectively learned by neural networks and whether the deep learning will be able to extract the knowledge from the approximated decision systems. Our results show that granulated datasets perform well when mined by deep learning tools. We have performed exemplary experiments using data from the UCI repository—Pytorch and Tensorflow libraries were used for building neural network and classification process. It turns out that deep learning method works effectively based on reduced training sets. Approximation of decision systems before neural networks learning can be important step to give the opportunity to learn in reasonable time.
One of the most popular families of techniques to boost classification are Ensemble methods. Random Forests, Bagging and Boosting are the most popular and widely used ones. This article presents a novel Ensemble Model, named Random Granular Reflections. The algorithm used in this new approach creates an ensemble of homogeneous granular decision systems. The first step of the learning process is to take the training system and cover it with random homogeneous granules (groups of objects from the same decision class that are as little indiscernible from each other as possible). Next, granular reflection is created, which is finally used in the classification process. Results obtained by our initial experiments show that this approach is promising and comparable with other tested methods. The main advantage of our new method is that it is not necessary to search for optimal parameters while looking for granular reflections in the subsequent iterations of our ensemble model.
In the era of Big data, there is still place for techniques which reduce the data size with maintenance of its internal knowledge. This problem is the main subject of research of a family of granulation techniques proposed by Polkowski. In our recent works, we have developed new, really effective and simple techniques for decision approximation, homogenous granulation and epsilon homogenous granulation. The real problem in this family of methods was the choice of an effective parameter of approximation for any datasets. It was resolved by homogenous techniques. There is no need to estimate the optimal parameters of approximation for these methods, because those are set in a dynamic way according to the data internal indiscernibility level. In this work, we have presented an extension of the work presented at ICIST 2018 conference. We present results for homogenous and epsilon homogenous granulation with the comparison of its effectiveness.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
customersupport@researchsolutions.com
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.