prikaz prve stranice dokumenta Naučene indeksne strukture
No public access
undergraduate thesis
Naučene indeksne strukture
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. urn:nbn:hr:168:253205
Paulinović, Mate
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Paulinović, M. (2018). Naučene indeksne strukture (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:253205

Paulinović, Mate. "Naučene indeksne strukture." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:253205

Paulinović, Mate. "Naučene indeksne strukture." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2018. https://urn.nsk.hr/urn:nbn:hr:168:253205

Paulinović, M. (2018). 'Naučene indeksne strukture', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 24 December 2024, https://urn.nsk.hr/urn:nbn:hr:168:253205

Paulinović M. Naučene indeksne strukture [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2018 [cited 2024 December 24] Available at: https://urn.nsk.hr/urn:nbn:hr:168:253205

M. Paulinović, "Naučene indeksne strukture", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2018. Available at: https://urn.nsk.hr/urn:nbn:hr:168:253205

Metadata
TitleNaučene indeksne strukture
Title (english)Task: Learned index structures
AuthorMate Paulinović
MentorMile Šikić (mentor)
Committee memberMile Šikić (predsjednik povjerenstva)
Committee memberKrešimir Križanović (član povjerenstva)
Committee memberRobert Vaser (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2018-09-11, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Tradicionalne strukture podataka za indeksiranje mogu se poboljšati metodama strojnog učenja. Neuronske mreže za tu primjenu uče distribuciju podataka spremljenih u indeksne strukture. Razvoj grafičkih kartica može povećati efikasnost i smanjiti vremena izvođenja takvih modela. U ovom radu pokazano je da naučeni model može bolje raspoređivati ključeve po tablici raspršenog adresiranja smanjujuću broj kolizija i memorijsko zauzeće. U radu je predstavljena teorijska podloga tablica raspršenog adresiranja i struktura modela strojnog učenja te implementacijski detalji razvijenog programskog koda.
Abstract (english)
Traditional index data structures can be improved using machine learning. Neural networks used for this learn the underlying data distributions of the data which is stored in these data structures. The development of graphics processing units could improve the efficiency of such models and reduce their execution time. It is shown in this thesis that a model can be learned which distributes the keys over the hash table more efficiently, thereby reducing the number of collisions as well as reducing the memory footprint. This thesis presents the theoretical background of hash tables and the machine learning structure as well as the implementation details of the developed source code.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:253205
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 on2019-03-13 23:39:16