prikaz prve stranice dokumenta Pametni sustav za predviđanje gledanosti TV kanala
Access restricted to authenticated users
master's thesis
Pametni sustav za predviđanje gledanosti TV kanala
Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. urn:nbn:hr:179:613325
Miletić, Dino
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Institutional repository: Repository of the Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture

Cite this document

Miletić, D. (2021). Pametni sustav za predviđanje gledanosti TV kanala (Master's thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:613325

Miletić, Dino. "Pametni sustav za predviđanje gledanosti TV kanala." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:613325

Miletić, Dino. "Pametni sustav za predviđanje gledanosti TV kanala." Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2021. https://urn.nsk.hr/urn:nbn:hr:179:613325

Miletić, D. (2021). 'Pametni sustav za predviđanje gledanosti TV kanala', Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 30 December 2024, https://urn.nsk.hr/urn:nbn:hr:179:613325

Miletić D. Pametni sustav za predviđanje gledanosti TV kanala [Master's thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2021 [cited 2024 December 30] Available at: https://urn.nsk.hr/urn:nbn:hr:179:613325

D. Miletić, "Pametni sustav za predviđanje gledanosti TV kanala", Master's thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2021. Available at: https://urn.nsk.hr/urn:nbn:hr:179:613325

Metadata
TitlePametni sustav za predviđanje gledanosti TV kanala
Title (english)Intelligent system for TV viewership forecasting
AuthorDino Miletić
MentorLjiljana Šerić (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2021-07-16, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovom radu opisan je postupak izrade sustava za predviđanje TV gledanosti. Predviđanje se temelji na primjeni algoritama strojnog učenja na prikupljenim podacima o TV gledanosti. Prikupljeni su stvarni podaci o gledanosti pohranjeni u zapisima sustava kojeg održava tvrtka Odašiljači i veze d.o.o. kao i podaci o TV rasporedu. Podaci su agregirani, obrađeni i transformirani u oblik pogodan za primjenu algoritama strojnog učenja. Proveden je cijeli niz eksperimenata kojima su testirane uspješnosti raznih pogodnih algoritama. Prvi dio algoritama se odnosi na regresijske metode, a drugi se odnosi na LSTM neuralnu mrežu. Za periodične trendove korišten je alat Prophet. Svaka od navedenih metoda predviđanje čini robusnijim i prilagodljivijim, pa se predlaže korištenje svih metoda kako bi predviđanja bila što uspješnija.
Abstract (english)
This thesis describes the procedure how to build a system for forecasting TV viewership. Prediction is performed by applying machine learning algorithms on TV viewership data. Real viewership data, as well as TV schedule dana, was collected from the dana store system maintened by company Odašiljači i veze d.o.o. Data were aggregated, parsed and transformed into shape that is suitable for application of machine learning algorithms. A whole series of experiments was carried out and performance was tested for each experiment. The first part of algorithms is based on regression and the second part is based on LSMT neural network. For periodic trends we used Prophet tool. Each of these methods makes forecasting more robust and adoptable, so it is suggested to use combination of above methods.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:613325
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 conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2021-11-25 14:28:00