Original title: Analýza nehodových lokalít a vzorcov dopravných nehôd
Translated title: Analysis of Accident Locations and Traffic Accident Patterns
Authors: Dubrovin, Maksim ; Hynek, Jiří (referee) ; Ondrušková, Magdaléna (advisor)
Document type: Bachelor's theses
Year: 2026
Language: eng
Publisher: Vysoké učení technické v Brně. Fakulta informačních technologií
Abstract: [eng] [cze]
Tato prace se zabyva analyzou dopravnich dat se zamerenim na identifikaci nehodovych lokalit a vzorcu dopravnich nehod. Nejprve jsou predstaveny principy analyzy dat, predikce a metody strojoveho uceni vhodne pro klasifikaci, shlukovani a identifikaci vzorcu v dopravnich datech. Nasledne jsou analyzovany dostupne datove zdroje, jako jsou otevrena data Policie Ceske republiky nebo komunitni data sluzby Waze, a jejich specifika z hlediska struktury a kvality. Prace se dale venuje analyze existujici aplikace pro zpracovani dopravnich a policejnich dat a navrhuje jeji rozsireni o modul pro analyzu nehodovych lokalit. Vysledkem je implementace rozsireneho reseni umoznujici identifikaci rizikovych mist, vizualizaci dat a vyhodnoceni vzorcu nehod. Soucasti prace je take testovani navrzene implementace a zhodnoceni dosazenych vysledku. This thesis focuses on the analysis of traffic data with an emphasis on identifying accident locations and traffic accident patterns. It first introduces the principles of data analysis, prediction, and machine learning methods suitable for classification, clustering, and the discovery of patterns in traffic datasets. Subsequently, the available data sources—such as open data from the Czech Police and community-generated data from Waze—are examined with regard to their structure and quality. The thesis also analyzes an existing application for processing traffic and police data and proposes its extension with a module for accident location analysis. The resulting solution includes the implementation of this extended functionality, enabling the identification of high-risk areas, data visualization, and the evaluation of accident patterns. The thesis concludes with testing of the implemented solution and an assessment of the achieved results.
Keywords: analyza dopravnich dat; dopravni analytika; dopravni nehody; geograficka analyza; klasifikace; nehodove lokality; policejni data; predikce; shlukovani; vizualizace dat; vzorce chovani; Waze data; accident hotspots; behavior patterns; classification; clustering; data visualization; geospatial analysis; police data; prediction; traffic accidents; traffic analytics; traffic data analysis; Waze data

Institution: Brno University of Technology (web)
Document availability information: Fulltext is available in the Brno University of Technology Digital Library.
Original record: http://hdl.handle.net/11012/258781

Permalink: http://www.nusl.cz/ntk/nusl-775918


The record appears in these collections:
Universities and colleges > Public universities > Brno University of Technology
Academic theses (ETDs) > Bachelor's theses