prikaz prve stranice dokumenta Metode redukcije dimenzije visokodimenzionalnih podataka
  Download
PDF 881.42 KB
undergraduate thesis
Metode redukcije dimenzije visokodimenzionalnih podataka
2020. urn:nbn:hr:126:464083
Prusina, Tomislav
Josip Juraj Strossmayer University of Osijek
Department of Mathematics
Chair of Applied Mathematics
Computer Science Research Group

Cite this document

Prusina, T. (2020). Metode redukcije dimenzije visokodimenzionalnih podataka (Undergraduate thesis). Retrieved from https://urn.nsk.hr/urn:nbn:hr:126:464083

Prusina, Tomislav. "Metode redukcije dimenzije visokodimenzionalnih podataka." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, 2020. https://urn.nsk.hr/urn:nbn:hr:126:464083

Prusina, Tomislav. "Metode redukcije dimenzije visokodimenzionalnih podataka." Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, 2020. https://urn.nsk.hr/urn:nbn:hr:126:464083

Prusina, T. (2020). 'Metode redukcije dimenzije visokodimenzionalnih podataka', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, accessed 16 May 2022, https://urn.nsk.hr/urn:nbn:hr:126:464083

Prusina T. Metode redukcije dimenzije visokodimenzionalnih podataka [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Department of Mathematics; 2020 [cited 2022 May 16] Available at: https://urn.nsk.hr/urn:nbn:hr:126:464083

T. Prusina, "Metode redukcije dimenzije visokodimenzionalnih podataka", Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Department of Mathematics, Osijek, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:126:464083

Please login to the repository to save this object to your list.
Metadata
TitleMetode redukcije dimenzije visokodimenzionalnih podataka
Title (english)Dimensionality Reduction Methods for High-Dimensional Data
AuthorTomislav Prusina
MentorDomagoj Matijević (mentor)
GranterJosip Juraj Strossmayer University of Osijek
Department of Mathematics
(Chair of Applied Mathematics)
(Computer Science Research Group)
Osijek
Defense date and country2020-09-29, Croatia
Scientific / art field,
discipline and subdiscipline		NATURAL SCIENCES
Mathematics
Other Mathematical Disciplines
Abstract
U ovome radu ćemo promatrati podatke u visokodimenzionalnim prostorima. Analiza podataka u takvim prostorima može biti računski i vremenski zahtjevna. Stoga uvodimo pojam redukcije (smanjenja) dimenzije kako bismo mogli lakše i brže provesti analizu te ćemo promatrati sljedeće metode za redukcije dimenzije: • slučajna projekcija, • analiza glavnih komponenti, • skaliranje najmanjih kvadrata. Za svaku metodu ćemo navesti prednosti i nedostatke te ćemo za metodu skaliranja najmanjih kvadrata dodatno proučiti algoritam pod nazivom "skaliranje majorizacijom komplicirane funkcije".
Abstract (english)
In this paper, we will observe data in high-dimensional spaces. Data analysis in such spaces can be a computational and time demanding. Therefore, we introduce the concept of dimensionality reduction so that we can more easily and quickly perform data analysis. We will observe the following three methods for reducing dimensions: • random projection, • principal component analysis, • least-squares scaling. For each method we will list the advantages. Additionaly, for the least-squares scaling method we will study the algorithm: scaling by majorizing a complicated function.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:126:464083
Study programmeTitle: University undergraduate study programme in mathematics
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) matematike (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) matematike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2020-10-30 08:23:27