Assessment of the Impact of Traffic Police Preventive Interventions
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F46747885%3A24310%2F17%3A00004208" target="_blank" >RIV/46747885:24310/17:00004208 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Assessment of the Impact of Traffic Police Preventive Interventions
Original language description
Free publicly available datasets describing weather and traffic accidents in the Czech Republic have been used for the training of a feed-forward neural network so that it could predict the level of the number of traffic accidents and their cost from weather, weekday, and month. The neural network learns the data for each of the 14 Czech regions separately and also the idea of cross-validation is utilized. The data for each learning task have been separated into Training Set, Development Test Set, and Test Set. Then a statistically significant number of experiments with neural network to get the accuracy of the prediction of the Test Set that happens when the accuracy of the Development Test Set is maximized have been conducted. The aim of the research is to learn whether this methodology can statistically detect any significant difference of the accuracy of prediction between the Test Set formed from days with interventions reported by the Czech Police and the randomly selected Test Set using the assumption that the neural network learns dependencies not affected by preventive interventions.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
—
OECD FORD branch
10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)
Result continuities
Project
—
Continuities
I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace
Others
Publication year
2017
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Mathematical Methods in Economics MME 2017
ISBN
978-80-7435-678-0
ISSN
—
e-ISSN
—
Number of pages
6
Pages from-to
505-510
Publisher name
Gaudeamus, University of Hradec Králové
Place of publication
Hradec Králové
Event location
Hradec Králové
Event date
Jan 1, 2017
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
000427151400086