Klasifikacija virusnih zaporedij s strojnim učenjem

diplomsko delo

Matej Kopar (Author), Tomaž Curk (Mentor)

Abstract

V diplomskem delu smo uporabili različne metode strojnega učenja za uvrščanje virusnih zaporedij v ustrezne taksonomske skupine. Z dostopanjem do podatkovne zbirke NCBI, ki hrani biološke in biotehnološke podatke, smo najprej sestavili celotno taksonomsko strukturo znanih virusnih zaporedij. Podatke smo ustrezno filtrirali in tako zgradili množico učnih primerov. Nato smo uporabili klasične metode strojnega učenja in metodo strukturiranega napovedovanja in ovrednotili uspešnost napovedovanja v taksonomske skupine. V delu smo preučili, kateri načini opisovanja genomskih zaporedij so najprimernejši. Opis genomskih zaporedij s k-terkami ne zajame vseh podrobnosti genomov, zato so najboljši doseženi rezultati le nekoliko boljši od večinskega klasifikatorja. Predznanje o evolucijski povezanosti taksonomskih skupin nekoliko izboljša napovedi modelov, ki to znanje lahko uporabijo.

Keywords

strojno učenje;klasifikacija;metoda podpornih vektorjev;naključni gozdovi;virusna zaporedja;strukturirano strojno učenje;računalništvo;računalništvo in informatika;univerzitetni študij;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2015
Typology:	2.11 - Undergraduate Thesis
Organization:	UL FRI - Faculty of Computer and Information Science
Publisher:	[M. Kopar]
UDC:	004.85(043.2)
COBISS:	1536600003
Views:	821
Downloads:	121
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Classification of viral genomes using machine learning
Secondary abstract:	In this diploma thesis our goal was to classify viral sequences into taxonomic groups by using different machine learning methods. We assembled the taxonomic structure by collecting data from NCBI web site. To clean the data we applied several filtering steps. We then evaluated the predictive performance of classical and structured machine learning methods on the task of classification in taxonomy groups. We wanted to determine the most suitable way to describe genomic sequences. Using k-mers to describe the genomic composition yielded poor predictive models, with best performance slightly above the performance of the majority classifier. Methods, which are able to use prior knowledge on the taxonomic relationships between classes, performed slightly better than methods, which did not use such information.
Secondary keywords:	machine learning;classification;support vector machine;random forest;viral sequences;structured machine learning;computer science;computer and information science;diploma;
File type:	application/pdf
Type (COBISS):	Bachelor thesis/paper
Study programme:	1000468
Embargo end date (OpenAIRE):	1970-01-01
Thesis comment:	Univ. v Ljubljani, Fak. za računalništvo in informatiko
Pages:	43 str.
ID:	9043477