Iskanje skritih informacij v meritvah kvalitete zraka s pomočjo podatkovnega rudarjenja

diplomsko delo

Vito Abeln (Author), Iztok Fister (Mentor)

Abstract

Diplomska naloga predstavlja postopek iskanja skritih podatkov iz meritev kvalitete zraka s pomočjo rudarjenja asociativnih pravil. Rudarjenje asociativnih pravil je tehnika, preko katere lahko pridobimo zanimive povezave med podatki iz večjih podatkovnih množic. V zaključnem delu smo prikazali postopek pridobivanja podatkov, obdelavo podatkov, razlago uporabljenih algoritmov in njihovo implementacijo. Opisali smo algoritme Apriori, ECLAT in Fp-growth, ki se uporabljajo pri asociativnem rudarjenju pravil. Predstavili smo tudi numerično rudarjenje asociativnih pravil, pri katerem smo uporabili algoritem optimizacije roja delcev. Rezultati rudarjenja so razkrili različne povezave med vrednostmi meritev, ki smo jih razložili in vizualizirali s pomočjo raznih grafov.

Keywords

rudarjenje asociativnih pravil;kvaliteta zraka;asociativna pravila;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2023
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[V. Abeln]
UDC:	004.62(043.2)
COBISS:	181668355
Views:	321
Downloads:	25
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Searching for hidden information of air quality measures with data mining
Secondary abstract:	This thesis presents a procedure for finding hidden data from air quality measurements using associtation rule mining. Association rule mining is a technique through which interesting associations between data from large datasets can be extracted. In this thesis we show the data mining process, data processing, explanation of the algorithms used and their implementation. We describe the Apriori, ECLAT and Fp-growth algorithms used in associative rule mining. We also presented numerical associative rule mining which used the particle swarm optimisation algorithm. The mining results revealed different relationships between the measurement values, which we explained and visualised using different graphs.
Secondary keywords:	association rule mining;air quality;association rules;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Informatika in tehnologije komuniciranja
Pages:	1 spletni vir (1 datoteka PDF (61 f.))
ID:	19965742