prikaz prve stranice dokumenta Razvoj sustava mikropredikcija za ponavljajuća predviđanja
No public access
master's thesis
Razvoj sustava mikropredikcija za ponavljajuća predviđanja
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. urn:nbn:hr:168:761654
Celinšćak, Antonio
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Celinšćak, A. (2024). Razvoj sustava mikropredikcija za ponavljajuća predviđanja (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:761654

Celinšćak, Antonio. "Razvoj sustava mikropredikcija za ponavljajuća predviđanja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:761654

Celinšćak, Antonio. "Razvoj sustava mikropredikcija za ponavljajuća predviđanja." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2024. https://urn.nsk.hr/urn:nbn:hr:168:761654

Celinšćak, A. (2024). 'Razvoj sustava mikropredikcija za ponavljajuća predviđanja', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 06 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:761654

Celinšćak A. Razvoj sustava mikropredikcija za ponavljajuća predviđanja [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2024 [cited 2024 November 06] Available at: https://urn.nsk.hr/urn:nbn:hr:168:761654

A. Celinšćak, "Razvoj sustava mikropredikcija za ponavljajuća predviđanja", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:168:761654

Metadata
TitleRazvoj sustava mikropredikcija za ponavljajuća predviđanja
Title (english)Creating a micropredictions oracle for repeated forecast tasks
AuthorAntonio Celinšćak
MentorMario Brčić (mentor)
Committee memberMario Brčić (predsjednik povjerenstva)
Committee memberMirjana Domazet-Lošo (član povjerenstva)
Committee memberNikica Hlupić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2024-07-05, Croatia
Abstract
Od početka vremena ljudi su pokušavali predvidjeti budućnost raznim tehnikama i vjerovanjima. Danas to rade korištenjem umjetne inteligencije i složenih matematičkih modela. Jedan od najnovijih pristupa predviđanju je sustav mikropredikcija koji je osmislio Peter Cotton. Autor za sustav mikropredikcija kaže da će s vremenom davati najbolja rješenja za djelić danas potrebne cijene. Sustav mikropredikcija sastoji se od mreže tokova podataka po kojoj putuju algoritmi koji pokušavaju riješiti probleme povezane s tim podacima. Svaki zadatak rješava se natjecanjem algoritama motiviranih nagradom, a natjecanja vodi i organizira o-stroj. Ovaj rad pokazuje izradu jednog o-stroja za zadatak predviđanja broja noćenja u skloništu u Calgaryju. U natjecanju sudjeluje šest različitih vrsta modela i cijeli sustav se testira različitim kombinacijama parametara da se dobe zaključci o njihovom utjecaju i o radu o-stroja.
Abstract (english)
Since the beginning of time, people have tried to predict the future with various techniques and beliefs. Today, they do this using artificial intelligence and complex mathematical models. One of the most recent approaches to forecasting is the microprediction system devised by Peter Cotton. The author of the microprediction system says that over time, it will provide the best solutions for a fraction of the price required today. A microprediction system consists of a network of data streams along which algorithms travel and attempt to solve problems associated with that data. A competition of prize-motivated algorithms solves each task, and the competitions are run and organized by Oracle. This paper demonstrates the creation of an Oracle to predict the number of overnight stays in a Calgary shelter. Six different types of models participate in the competition and the entire system is tested with varying combinations of parameters to draw conclusions about their effects and the operation of Oracle.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:761654
Study programmeTitle: Computing
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2024-10-12 20:16:40