The recent progress in information and communication technologies has enabled us to obtain and store very large amounts of data. The main problem is how to find and extract useful information contained in data. The main goal of this study is to assess the possibility of using Adaptive resonance theory (ART) neural networks for cluster analysis and information retrieval. Their properties, behavior and success rate for different types of artificial and real data were determined as well as optimal values of their parameters for this purpose. Based on achieved results, the possibility of using ART neural networks for Big data analysis was assessed. Then the application based on ART principles with included graphical user interface was implemented and this process was described.
| Identifer | oai:union.ndltd.org:nusl.cz/oai:invenio.nusl.cz:251328 |
| Date | January 2015 |
| Creators | Janů, Ondřej |
| Source Sets | Czech ETDs |
| Language | Czech |
| Detected Language | English |
| Type | info:eu-repo/semantics/masterThesis |
| Rights | info:eu-repo/semantics/restrictedAccess |
Page generated in 0.0016 seconds