Please use this identifier to cite or link to this item:
Title: Combining probabilistic neural networks and decision trees for maximally accurate and efficient accident prediction
Authors: Tambouratzis, Tatiana 
Souliou, Dora 
Chalikias, Miltiadis 
Gregoriades, Andreas 
Keywords: Accident prediction;Accident severity;Data sets;Input parameter;Output parameters;Probabilistic neural networks
Category: Computer and Information Sciences
Field: Engineering and Technology
Issue Date: 2010
Source: Proceedings of the International Joint Conference on Neural Networks 2010, Article number 5596610 2010 6th IEEE World Congress on Computational Intelligence, WCCI 2010 - 2010 International Joint Conference on Neural Networks, IJCNN 2010; Barcelona; Spain; 18 July 2010 through 23 July 2010; Category numberCFP10IJS-DVD; Code 85188
Journal: Proceedings of the International Joint Conference on Neural Networks 
Conference: 6th IEEE World Congress on Computational Intelligence 
Abstract: The extent to which accident severity can be predicted from accident-related data collected at a variety of locations is investigated. The 2005 accident dataset brought together by the Republic of Cyprus Police is employed; this dataset comprises 1407 records of 43 continuous and categorical input parameters and a single categorical output parameter representing accident severity. No transformation of the database has been opted for, either by extracting the parameters that are significant for the prediction task or by modifying the records in any way (e.g. via record selection or transformation). Aiming at maximally accurate and efficient prediction, a combination of probabilistic neural networks (PNN's) and decision trees (DT's) is implemented: the simple training and direct operation of the PNN is complemented by the hierarchical, exhaustive and recursive construction of the DT. By training pairs of PNN's on data from the partitions derived from the minimal necessary number of top DT nodes, both efficiency and accident prediction accuracy are maximized. © 2010 IEEE.
ISBN: 9781424469178
ISSN: 2-s2.0-79959484929
DOI: 10.1109/IJCNN.2010.5596610
Rights: © 2010 IEEE.
Type: Conference Papers
Appears in Collections:Δημοσιεύσεις σε συνέδρια /Conference papers - poster -presentation

Show full item record

Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.