Thesis Details
Fúze dat pro klasifikaci síťových zařízení
This work is focused on solving information fusion when dealing with multiple data sources in computer network monitoring. A solution built on the concept of classification rules configured by experts is presented. Configuration is simplified using a designated configuration language interpreted by the solution. The classification rules enable coverage of diverse types of data. The result is given as a label from specified taxonomy. Using a taxonomy maintains the different levels of detail between the data sources, even in the output label. The solution also uses the Dempster-Schafer theory for merging labels from different sources into a single output label. Results of experiments show that information fusion in this context does increase the accuracy of device classification. A process of rule optimization was developed based on testing and experiments with a dataset from a real network. The accuracy was increased by 19 % compared to the original solution using this process.
Dempster-Schafer theory, Transferable Belief Model, network devices, network assets, network monitoring, OS classification, device type classification, ADiCT
Hradiš Michal, Ing., Ph.D. (DCGM FIT BUT), člen
Jaroš Jiří, doc. Ing., Ph.D. (DCSY FIT BUT), člen
Křivka Zbyněk, Ing., Ph.D. (DIFS FIT BUT), člen
Lengál Ondřej, Ing., Ph.D. (DITS FIT BUT), člen
@bachelorsthesis{FITBT25057, author = "Ond\v{r}ej Sedl\'{a}\v{c}ek", type = "Bachelor's thesis", title = "F\'{u}ze dat pro klasifikaci s\'{i}\v{t}ov\'{y}ch za\v{r}\'{i}zen\'{i}", school = "Brno University of Technology, Faculty of Information Technology", year = 2022, location = "Brno, CZ", language = "czech", url = "https://www.fit.vut.cz/study/thesis/25057/" }