Ocenjevanje zanesljivosti posameznih klasifikacij z lokalnimi metodami

Povzetek

Ključne besede

Podatki

Jezik:	Slovenski jezik
Leto izida:	2009
Tipologija:	2.11 - Diplomsko delo
Organizacija:	UL FRI - Fakulteta za računalništvo in informatiko
Založnik:	[D. Pevec]
UDK:	004(043.2)
COBISS:	7296596
Št. ogledov:	1137
Št. prenosov:	238
Ocena:	0 (0 glasov)
Metapodatki:

Ostali podatki

Sekundarni jezik:	Angleški jezik
Sekundarni naslov:	[Evaluating reliability of individual classifications with local methods]
Sekundarni povzetek:	In this thesis we take upon different approaches for estimating reliability of individual classification predictions made by classifiers based on supervised learning. The general definition of the term reliability is the ability to perform required functions under stated conditions. In machine learning, we refer to accuracy, as in the ability to provide accurate predictions. We face the problem that measures of reliability are not quantitatively defined. We can therefore only conceive estimates. Reliability estimates of individual predictions provide valuable information that could be beneficial in individual predictions assessment of use. For the needs of our thesis we develop several methods for reliability estimation based on existing approaches of local methods and the variance of a bagged model. We test our methods on various available real-life and artificial datasets and compare our methods with those based on inverse transduction. Methods were tested on 20 different datasets on 7 classification models and the estimates were calculated using 11 measures of similarity. We applied three statistical methods to our results. We came to a conclusion that these tests do not give clear results, as Q-Q plots only vaguely support calculated correlation. Correlation tests show potential of estimates LCV and BAGV as they demonstrated best on average performance. Second-comers with good result were estimates TRANS1 and CNK, while other estimates failed to excel.
Sekundarne ključne besede:	machine learning;classification;reliability estimates;classification accuracy;computer science;diploma;
Vrsta datoteke:	application/pdf
Vrsta dela (COBISS):	Diplomsko delo
Komentar na gradivo:	Univerza v Ljubljani, Fakulteta za računalništvo in informatiko
Strani:	86 str.
ID:	23829080