Fakultet prometnih znanosti u suradnji sa Znanstvenim centrom izvrsnosti za znanost u podatcima i kooperativne sustave te pripadnim projektom Napredne metode i tehnologije u znanosti o podatcima i kooperativnim sustavima (DATACROSS) i udrugom ITS Hrvatska pozivaju Vas na predavanje:
“Estimating motorized travel mode choice using classifiers: An application for high-dimensional multicollinear data”
koje će održati profesor André Luiz Cunha, University of São Paulo (USP), São Carlos School of Engineering (EESC), Brazil u četvrtak, 07. lipnja 2018. godine u dvorani D1 na Fakultetu prometnih znanosti Sveučilište u Zagrebu, Vukelićeva 4 s početkom u 13.00 h. Predavanje će se prenositi i preko weba preko poveznice https://connect.carnet.hr/fpz_predavanje/.
Predavanje će biti na engleskom jeziku. Predviđeno trajanje s raspravom je 90 minuta te je otvoreno za sve zainteresirane, a posebice studente.
Više o predavanju i predavaču pročitajte u nastavku obavijesti.
Abstract: Studies in the field of discrete choice analysis are crucial for transportation planning. Generally, travel demand models are based on the maximization of the random utility and straightforward mathematical functions, such as logit models. These assumptions lead to a continuous model that presents constraints concerning fitting the data. Artificial Neural Networks (ANN) and Classification Trees (CT) are classification techniques that can be applied to discrete choice models. These techniques can overcome some disadvantages of traditional modeling, especially the drawback of not being able to model high-dimensional multicollinear data. We will present the outcomes of the performance comparison of estimating motorized travel mode choice through ANN and CT with a binary logit in a multicollinear study case (aggregated and disaggregated covariates): Origin-Destination Survey carried out in São Paulo Metropolitan Area, Brazil in 2007. Classification techniques have shown a good ability to forecast, as well as to recognize travel behavior patterns.
CV: Andre Luiz Cunha is a PhD. Professor at the University of São Paulo (USP), São Carlos School of Engineering (EESC), in Brazil. He holds a degree in Civil Engineering from Federal University of Mato Grosso do Sul (UFMS) in 2004. He received the M.Sc. in Transport Engineering from USP in 2007 and the Ph.D. in Transportation Engineering from USP in 2013. Since 2014, he has been actively contributing to teaching and research in programming, database and data analysis at Department of Transport Engineering, USP-EESC. His research interest is in Intelligent Transportation Systems (ITS) with focus in use of sensors and development of devices to collect traffic data in field, computer vision, anomalies detection, pattern recognition, and traffic simulation.