Combining probabilistic neural networks and decision trees for maximally accurate and efficient accident prediction

Tambouratzis, T; Souliou, D; Chalikias, M; Gregoriades, A

dc.contributor.author	Tambouratzis, T	en
dc.contributor.author	Souliou, D	en
dc.contributor.author	Chalikias, M	en
dc.contributor.author	Gregoriades, A	en
dc.date.accessioned	2014-03-01T02:46:43Z
dc.date.available	2014-03-01T02:46:43Z
dc.date.issued	2010	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/32806
dc.subject	Prediction Accuracy	en
dc.subject	Probabilistic Neural Network	en
dc.subject	Relational Data	en
dc.subject	Decision Tree	en
dc.subject.other	Accident prediction	en
dc.subject.other	Accident severity	en
dc.subject.other	Data sets	en
dc.subject.other	Input parameter	en
dc.subject.other	Output parameters	en
dc.subject.other	Prediction tasks	en
dc.subject.other	Probabilistic neural networks	en
dc.subject.other	Recursive construction	en
dc.subject.other	Accidents	en
dc.subject.other	Decision trees	en
dc.subject.other	Forecasting	en
dc.subject.other	Plant extracts	en
dc.subject.other	Trees (mathematics)	en
dc.subject.other	Neural networks	en
dc.title	Combining probabilistic neural networks and decision trees for maximally accurate and efficient accident prediction	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/IJCNN.2010.5596610	en
heal.identifier.secondary	5596610	en
heal.identifier.secondary	http://dx.doi.org/10.1109/IJCNN.2010.5596610	en
heal.publicationDate	2010	en
heal.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.	en
heal.journalName	Proceedings of the International Joint Conference on Neural Networks	en
dc.identifier.doi	10.1109/IJCNN.2010.5596610	en