Strojno učenje prepoznavanja rukopisne numeracije

Slota, Marko

undergraduate thesis

Strojno učenje prepoznavanja rukopisne numeracije

Zagreb: University of Zagreb, University of Zagreb, Faculty of Humanities and Social Sciences, 2020. urn:nbn:hr:131:837566

Slota, Marko

University of Zagreb
Faculty of Humanities and Social Sciences
Department of information and Communication sciences

Institutional repository: ODRAZ

Cite this document

APA 6th Edition

Slota, M. (2020). Strojno učenje prepoznavanja rukopisne numeracije (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Humanities and Social Sciences. Retrieved from https://urn.nsk.hr/urn:nbn:hr:131:837566

MLA 8th Edition

Slota, Marko. "Strojno učenje prepoznavanja rukopisne numeracije." Undergraduate thesis, University of Zagreb, Faculty of Humanities and Social Sciences, 2020. https://urn.nsk.hr/urn:nbn:hr:131:837566

Chicago 17th Edition

Harvard

Slota, M. (2020). 'Strojno učenje prepoznavanja rukopisne numeracije', Undergraduate thesis, University of Zagreb, Faculty of Humanities and Social Sciences, accessed 28 April 2024, https://urn.nsk.hr/urn:nbn:hr:131:837566

Vancouver

Slota M. Strojno učenje prepoznavanja rukopisne numeracije [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Humanities and Social Sciences; 2020 [cited 2024 April 28] Available at: https://urn.nsk.hr/urn:nbn:hr:131:837566

IEEE

M. Slota, "Strojno učenje prepoznavanja rukopisne numeracije", Undergraduate thesis, University of Zagreb, Faculty of Humanities and Social Sciences, Zagreb, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:131:837566

Cite this item: https://urn.nsk.hr/urn:nbn:hr:131:837566

Metadata

Title	Strojno učenje prepoznavanja rukopisne numeracije
Title (english)	Handwritten digit recognition using machine learning
Author	Marko Slota
Mentor	Ivan Dunđer (mentor)
Mentor	Sanja Seljan (sumentor)
Committee member	Vjera Lopina (predsjednik povjerenstva)
Committee member	Ivan Dunđer (član povjerenstva)
Committee member	Sanja Seljan (član povjerenstva)
Granter	University of Zagreb Faculty of Humanities and Social Sciences (Department of information and Communication sciences) Zagreb
Defense date and country	2020-09-18, Croatia
Scientific / art field, discipline and subdiscipline	SOCIAL SCIENCES Information and Communication Sciences
Abstract	Završni rad na temu strojnog učenja prepoznavanja rukopisne numeracije detaljno prikazuje podjelu algoritama zaduženih za strojno učenje na način da vrši uvid u same pristupe strojnom učenju i načine na koje se ti pristupi mogu koristiti kada dođe do problematike prepoznavanja ručno napisanih brojčanih znamenaka. Svi navedeni pristupi nadalje su razgranati kako bi se postigao jasan pregled najčešćih oblika klasifikatora zaduženih za istraživanja na ovu temu. Ovaj rad daje uvid u postotak točnosti i pogreške dobivenih na osnovu istraživanja vršenih za svaki pojedini algoritam koji se koristi u obliku sustava za klasifikaciju korištenih ulaznih podataka. Svi rezultati dobiveni testiranjem navedenih algoritama tablično su prikazani i detaljno objašnjeni. Također, vršen je i prikaz eventualnih problema koji se događaju pri stvaranju predikcija o kojoj se ručno napisanoj znamenci uistinu radi, kao i koja su eventualna rješenja za iste. U završnome dijelu rada prikazano je softwaresko rješenje ta temu ove problematike u obliku koda napisanog u programskom jeziku Python.
Abstract (english)	The final work on theme of handwritten digit recognition using machine learning shows in detail the division of machine learning algorithms by insight into the machine learning approaches and ways those approaches can be used in problems related to handwritten numbering. All of the above approaches are further explained in order to achieve a clear overview of the most common forms of classifiers in charge of research on this topic. This paper provides insight into the percentage of accuracy and error obtained based on the research performed for each individual algorithm it uses in the form of a system for classifying the input data used. All results obtained by testing this algorithm are tabulated and published in detail. Also, a presentation of possible problems that occur when creating a prediction about which handwritten digit is truly used, as well as possible solutions for the same. The final part of the paper presents a software solution on the topic in the form of code written in the Python programming language.
Keywords
Keywords (english)
Language	croatian
URN:NBN	urn:nbn:hr:131:837566
Study programme	Title: Information Sciences Study programme type: university Study level: undergraduate Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informacijskih znanosti (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informacijskih znanosti)
Type of resource	Text
File origin	Born digital
Access conditions	Open access
Terms of use
Created on	2020-09-20 19:36:55

Search form