prikaz prve stranice dokumenta Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja
No public access
undergraduate thesis
Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. urn:nbn:hr:168:359771
Vesinger, Dunja
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Vesinger, D. (2017). Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:359771

Vesinger, Dunja. "Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:359771

Vesinger, Dunja. "Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2017. https://urn.nsk.hr/urn:nbn:hr:168:359771

Vesinger, D. (2017). 'Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 22 July 2024, https://urn.nsk.hr/urn:nbn:hr:168:359771

Vesinger D. Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2017 [cited 2024 July 22] Available at: https://urn.nsk.hr/urn:nbn:hr:168:359771

D. Vesinger, "Klasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2017. Available at: https://urn.nsk.hr/urn:nbn:hr:168:359771

Metadata
TitleKlasifikacija rukom pisanih slova uporabom ograničenog Boltzmannovog stroja
Title (english)Classification of Handwritten Alphabet Letters using Restricted Boltzmann Machine
AuthorDunja Vesinger
MentorMarko Čupić (mentor)
Committee memberMarko Čupić (predsjednik povjerenstva)
Committee memberMiroslav Popović (član povjerenstva)
Committee memberMartin Soldić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2017-07-10, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Ovaj rad proučava klasifikaciju rukom pisanih slova engleske abecede uporabom binarnog ograničenog Boltzmannovog stroja. Korištena metoda klasifikacije provodi se nad reprezentacijama slova u obliku binarnih vektora na koje je nadodan jednojedinični kod. Algoritmom kontrastne divergencije s jednim korakom nastoji se naučiti Boltzmannov stroj da generira odgovarajući jednojedinični kod za svaki primjerak binarizirane slike rukom pisanog slova. Metoda klasifikacije primijenjena je na skupu svih slova engleske abecede i ispravno klasificira samo 33.88% primjera. Na skupu primjera koji sadrži prvih šest slova engleske abecede dobiven je znatno bolji stupanj ispravne klasifikacije koji iznosi 57.54%.
Abstract (english)
This thesis studies the classification of handwritten letters of the English alphabet using a restricted Boltzmann machine. The method utilized in this paper uses binary representations of the images containing letters and adds a one-hot encoding to each vector. The contrastive divergence algorithm with one step is used to train the restricted Boltzmann machine to generate the proper one-hot encoding for each binary image of a letter. The method of classification used in this paper correctly classifies only 33.88% of examples when used on the training set containing all the letters of the English alphabet. On a subset containing the first six letters of the alphabet, it gives a significantly better accuracy of 57.54%.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:359771
Study programmeTitle: 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 resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2019-03-14 01:52:32