Merging Bayesian Networks based on different types of input knowledge
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21260%2F18%3A00326495" target="_blank" >RIV/68407700:21260/18:00326495 - isvavai.cz</a>
Result on the web
—
DOI - Digital Object Identifier
—
Alternative languages
Result language
angličtina
Original language name
Merging Bayesian Networks based on different types of input knowledge
Original language description
Bayesian networks are one of the most suitable tools for traffic accident data analysis. A well-designed and trained Bayesian network is the first assumption for obtaining good results. Two sources of information can be used for this purpose. They are information extracted from measured historical data and also information specified by an expert. The latter one can also involve some generally known rules or knowledge based on traffic regulations or safety rules. Mostly, only one of these sources of information is used. After training the network can be evaluated with standard methods based on likelihood or prediction error evaluation. The goal of this paper is to show that if two Bayesian networks are created, e.g. one from the data and second from the expert knowledge, they can be merged into one which joins the objective information from data with the subjective opinion from expert and which has better evaluation than the original ones.
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
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2018
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Young Transportation Engineers Conference 2018
ISBN
978-80-01-06464-1
ISSN
—
e-ISSN
—
Number of pages
9
Pages from-to
53-61
Publisher name
Fakulta dopravní
Place of publication
Praha
Event location
Praha, Horská 3
Event date
Nov 1, 2018
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—