Analiza podataka vezanih uz zračni promet

undergraduate thesis

• Title (english): Analyzing Air Traffic Data
• Author: Vedran Kolka
• Mentor: Damir Pintar (mentor)
• Committee member: Damir Pintar (predsjednik povjerenstva)
• Committee member: Luka Humski (član povjerenstva)
• Committee member: Frano Škopljanac-Mačina (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-07-09, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu je analiziran skup podataka o zračnom prometu na području Sjedinjenih Američkih Država. Prvo je provedena eksploratorna analiza kojom se dobiva uvid u karakteristike skupa s kojim se radi. Nakon eksploratorne analize, cilj je bio sagraditi nekoliko modela kojima bi se pokušalo predviđati cijene zrakoplovnih karata. Prvi isprobani model bio je linearni model.
Drugi model bio je dobiven metodom k-najbližih susjeda. Ova metoda se pokazala boljom od linearne regresije, ali još uvijek ne dovoljno zadovoljavajućom. Treći model bio je umjetna neuronska mreža sa zakašnjenjem koja je na ulaz primala 8 prethodnih cijena. Ovaj model je u usporedbi s prethodnim modelima bio znatno bolji, ali objektivno, prosječna apsolutna pogreška je i ovdje bila daleko od idealne.

Abstract (english)

In this thesis, a dataset regarding air traffic data in The United States of America is analysed. First, an exploratory analysis was conducted so as to gain an insight into the dataset characteristics.
After the exploratory analysis, the goal was to create a few models for predicting flight fares.
The first created model was a linear model. The second model was built with the k nearest neighbours method. This method proved to be better than linear regression, but still not quite satisfactory. The third model was a time delay neural network which was given 8 previous fares as its input. This model proved to be much better comparing to the previous two models, but objectively the mean absolute error was still far from ideal.

• Keywords: eksploratorna analiza; predviđanje; linearna regresija; k-najbližih susjeda; vremenske serije; umjetna neuronska mreža sa zakašnjenjem
• Keywords (english): exploratory analysis; prediction; linear regression; k-nearest neighbours; time series; time delay neural network
• Language: croatian
• URN:NBN: urn:nbn:hr:168:330967
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 09:07:44