Prediction of Daily Traffic Accident Counts and Related Economic Damage in the Czech Republic

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24310%2F16%3A00000281" target="_blank" >RIV/46747885:24310/16:00000281 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Prediction of Daily Traffic Accident Counts and Related Economic Damage in the Czech Republic

Popis výsledku v původním jazyce

Free datasets describing weather and traffic accidents in the Czech Republic have been used for the training of neural network that would predict the number of traffic accidents and the level of economic damage for a given day. The aim of the research is to find out whether there are enough statistical dependencies in the available data so that a practically usable predictor could be trained from them. The Pearson’s chi-squared test was used to select input attributes for the neural network. The selected attributes are month, day of week, temperature in two selected preceding days and in the current day, precipitation, and snow. The neural network has been trained on the daily data of the years 2009 till 2014 divided into training and development test sets. The accuracy of the network after this training on more recent days is higher than majority voting, which can motivate a future research.

Název v anglickém jazyce

Prediction of Daily Traffic Accident Counts and Related Economic Damage in the Czech Republic

Popis výsledku anglicky

Free datasets describing weather and traffic accidents in the Czech Republic have been used for the training of neural network that would predict the number of traffic accidents and the level of economic damage for a given day. The aim of the research is to find out whether there are enough statistical dependencies in the available data so that a practically usable predictor could be trained from them. The Pearson’s chi-squared test was used to select input attributes for the neural network. The selected attributes are month, day of week, temperature in two selected preceding days and in the current day, precipitation, and snow. The neural network has been trained on the daily data of the years 2009 till 2014 divided into training and development test sets. The accuracy of the network after this training on more recent days is higher than majority voting, which can motivate a future research.

Druh

D - Stať ve sborníku

CEP obor

BB - Aplikovaná statistika, operační výzkum

OECD FORD obor

—

Projekt

—

Návaznosti

I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2016

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

34th International Conference Mathematical Methods in Economics 2016 Conference Proceedings

ISBN

978-80-7494-296-9

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

606-611

Název nakladatele

Technical University of Liberec

Místo vydání

Liberec

Místo konání akce

Liberec

Datum konání akce

1. 1. 2016

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

000385239500104