Neural network based traffic prediction

master's thesis

• Title: Predviđanje količine prometa zasnovano na neuronskoj mreži
• Author: Matea Ravnjak
• Mentor: Edouard Ivanjko (mentor)
• Committee member: Štefica Mrvelj (predsjednik povjerenstva)
• Committee member: Sadko Mandžuka (član povjerenstva)
• Committee member: Hrvoje Gold (član povjerenstva)
• Granter: University of Zagreb; Faculty of Transport and Traffic Sciences; (Division of Intelligent Transport Systems and Logistics); (Department of Intelligent Transport Systems); Zagreb
• Defense date and country: 2015-09-24, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Traffic and Transport Technology; Postal and Telecommunications Traffic

Abstract

U ovom radu kreirana je neuronska mreža s ciljem predviđanja količine prometa na transportnoj mreži. Pri tome se misli na cestovni transport (broj vozila) te na telekomunikacijski promet, odnosno prijenos podataka. Na osnovu ulaznih varijabli prikupljenih tijekom određenog vremena, predviđa se broj vozila odnosno količina podatkovnog prometa za idući vremenski interval. Korištena je višeslojna struktura neuronske mreže, a rezultati pokazuju da je neuronska mreža sposobna prilikom učenja aproksimirati funkciju ulaznih i izlaznih varijabli uz visoku točnost.

Abstract (english)

In this paper, neural network is created in order to predict the amount of traffic in the transport networks. This refers to road transport (number of vehicles) and telecommunication traffic, i.e. data transfer. Based on the input variables collected during a certain time, neutral network made predictions for number of vehicles and the amount of data traffic for the next time interval. Multilayer structure neutral network is used, and the results show that the neural network is capable to learn approximate function of input and output variables with high accuracy.

• Keywords: umjetna inteligencija; neuronske mreže; predviđanje; transportni sustavi; višeslojna arhitektura; cestovna mreža; telekomunikacijska mreža
• Keywords (english): artificial intelligence; neural networks; prediction; transport systems; multi-layer architecture; road network; telecommunication network
• Language: croatian
• URN:NBN: urn:nbn:hr:119:749606
• Study programme: Title: Promet; smjerovi: Cestovni, Željeznički, Vodni, Zračni, Poštanski, Informacijsko-komunikacijski, Gradski; Study programme type: sveučilišni; Study level: diplomski; Academic / professional title: magistar/magistra inženjer/inženjerka prometa (magistar/magistra inženjer/inženjerka prometa)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2015-09-30 11:31:41