prikaz prve stranice dokumenta Predviđanje performansi interakcije pokaznom napravom
  Download
PDF 2.64 MB
master's thesis
Predviđanje performansi interakcije pokaznom napravom
2023. urn:nbn:hr:190:374789
Dudić, Mia
University of Rijeka
Faculty of Engineering

Cite this document

Dudić, M. (2023). Predviđanje performansi interakcije pokaznom napravom (Master's thesis). Rijeka: University of Rijeka, Faculty of Engineering. Retrieved from https://urn.nsk.hr/urn:nbn:hr:190:374789

Dudić, Mia. "Predviđanje performansi interakcije pokaznom napravom." Master's thesis, University of Rijeka, Faculty of Engineering, 2023. https://urn.nsk.hr/urn:nbn:hr:190:374789

Dudić, Mia. "Predviđanje performansi interakcije pokaznom napravom." Master's thesis, University of Rijeka, Faculty of Engineering, 2023. https://urn.nsk.hr/urn:nbn:hr:190:374789

Dudić, M. (2023). 'Predviđanje performansi interakcije pokaznom napravom', Master's thesis, University of Rijeka, Faculty of Engineering, accessed 09 July 2024, https://urn.nsk.hr/urn:nbn:hr:190:374789

Dudić M. Predviđanje performansi interakcije pokaznom napravom [Master's thesis]. Rijeka: University of Rijeka, Faculty of Engineering; 2023 [cited 2024 July 09] Available at: https://urn.nsk.hr/urn:nbn:hr:190:374789

M. Dudić, "Predviđanje performansi interakcije pokaznom napravom", Master's thesis, University of Rijeka, Faculty of Engineering, Rijeka, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:190:374789

Please login to the repository to save this object to your list.
Metadata
TitlePredviđanje performansi interakcije pokaznom napravom
Title (english)Predicting interaction performance with a pointing device
AuthorMia Dudić
MentorGoran Mauša (mentor)
MentorSandi Ljubić (komentor)
Committee memberMladen Tomić (predsjednik povjerenstva)
Committee memberSandi Ljubić (član povjerenstva)
Committee memberGoran Mauša (član povjerenstva)
GranterUniversity of Rijeka
Faculty of Engineering
Rijeka
Defense date and country2023-09, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
U ovome je radu napravljena desktop aplikacija za automatizirano testiranje korisnika prilikom korištenja pokaznog uređaja za rješavanje tunelskih zadataka. Prilikom testiranja mjerila se brzina rješavanja pojedinog zadatka. Zadatak se manifestira kao tunel kroz koji je potrebno proći odabranim interaktivnim uređajem. Konfiguracija tunela mijenja se u 15 različitih izvedba gdje se kombiniraju uvjeti težine tunela i različitih krivulja. Testiranje se sastojalo od 8 testnih slučajeva od kojih svaki sadrži 15 pojedinih zadataka. Za testne slučajeve postavljeni su uvjeti boje programa, idealne linije i glazbe. Također, prikupljene su dodatne osobne informacije za svakog korisnika. Napravljeno je testiranje na 50 korisnika te je napravljena baza podataka koja se dalje koristila za obradu i razvoj modela predviđanja. Za model predviđanja korišteni su algoritmi za rješavanje problema regresije te su njihovi rezultati dani u grafičkom i opisnom obliku. Korištena je unakrsna validacija kako bi se smanjio problem podtreniranosti i pretreniranosti modela. Za svaki su algoritam korištene metrike vrednovanja kako bi se definirala točnost modela. Algoritmi i njihovi rezultati potom su uspoređeni kako bi se točno prikazalo kojim je algoritmom dobiveno najbolje rješenje.
Abstract (english)
In this project, a desktop application was developed for automated user testing when using an interactive device for solving tunnel tasks. During testing, the speed of solving individual tasks was measured. The task is manifested as a tunnel that needs to be navigated through using an interactive device. The tunnel configuration varies in 15 different setups, combining tunnel difficulty conditions and different curves. Testing consisted of 8 test cases, each containing 15 individual tasks. Conditions for program color, ideal line, and music were set for the test cases. Additionally, personal information was collected for each user. Testing was conducted with 50 users, and a database was created, which was further used for data processing and model development. Regression algorithms were used for prediction modeling, and their results were presented in graphical and descriptive forms. Cross-validation was employed to mitigate the issues of underfitting and overfitting. Evaluation metrics were used for each algorithm to define the accuracy of the model. The algorithms and their results were then compared to accurately depict which algorithm produced the best solution.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:190:374789
Study programmeTitle: Computer Engineering
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 conditionsOpen access
Terms of useLicence Attribution 4.0 International (CC BY 4.0) icon
Created on2023-09-12 13:32:44