izpis_h1_title_alt

Uporaba predznanja pri vizualizaciji visokodimenzionalnih podatkov
ID US, PETER (Author), ID Demšar, Janez (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (1,40 MB)
MD5: 0B2136EF56BE35C3258541446D6B7314
PID: 20.500.12556/rul/47a68f96-9049-4caf-b056-c29c478563a2

Abstract
Vizualizacije podatkov so lahko odličen pristop k analizi in odkrivanju novega znanja o podatkih. O podatkih imamo velikokrat na voljo tudi določeno predznanje, ki se ga izplača uporabiti pri iskanju koristnih vizualizacij. V diplomski nalogi raziskujemo vpliv genskih skupin pridobljenih iz baze MSigDB na kakovost vizualizacij mikromrež DNA. Ukvarjamo se z vprašanjem, ali lahko skupine genov uporabimo za pridobitev boljših vizualizacij. V prvem poglavju predstavimo vizualizacijo podatkov na splošno in metodi razsevni diagram ter Radviz, ki se uporabljata tekom dela. Na koncu poglavja sledi predstavitev metode VizRank, ki nam omogoča avtomatsko ocenjevanje in iskanje kakovostnih vizualizacij. Drugo poglavje opiše podatke mikromrež in podatke o genskih skupinah. Sledi predstavitev našega eksperimentalnega dela, ki je razdeljeno v tri sklope. V vsakem sklopu predstavimo idejo, postopek, rešitve in analizo rezultatov eksperimentiranja. Zaključimo s predstavitvijo glavnih ugotovitev.

Language:Slovenian
Keywords:vizualizacije, VizRank, Radviz, razsevni diagram, mikromreže DNA
Work type:Bachelor thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2014
PID:20.500.12556/RUL-29466 This link opens in a new window
Publication date in RUL:15.09.2014
Views:1193
Downloads:301
Metadata:XML RDF-CHPDL DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Use of background knowledge in visualization of high-dimensional data
Abstract:
Data visualisation can be an extremely efficient way of analyzing and discovering new knowledge from data. More often than not, we have background knowledge about the data, which we can then use to find meaningful and useful visualizations. This thesis examines the influence of gene sets gained from MSigDB upon the quality of visualizations of DNA microarrays. We hypothesize, that we can use gene sets to gain better and clearer visualizations. In the first chapter we explain what data visualization is, and introduce our working methods. At the end of the first chapter we present a method called VizRank, which allows us to automatically find and rate quality visualizations. This is followed by the second chapter, in which we describe the DNA microarray data and the data of gene sets. In the last part we present our experiential work, which is split into three sections. In each individual section we present the idea, procedure, solution and analysis of the results of experimentation.

Keywords:visualisations, VizRank, Radviz, scatter plot, DNA microarrays

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back