Klasifikacija DNA sljedova korištenjem modela dubokog učenja

master's thesis

• Title (english): Deep-learning based DNA sequence classification
• Author: Jelena Matečić
• Mentor: Mirjana Domazet-Lošo (mentor)
• Committee member: Mirjana Domazet-Lošo (predsjednik povjerenstva)
• Committee member: Boris Milašinović (član povjerenstva)
• Committee member: Danijel Mlinarić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-04, Croatia

Abstract

U radu se objašnjava arhitektura transformera i motivacija za njezino korištenje u rješavanju problema klasifikacije DNA sljedova. Opisana je ostvarena arhitektura korištenjem programskog jezika Python. Opisane su dvije metode pripreme skupa podataka za treniranje modela: k-mer i label encoding. Uspoređeni su rezultati varijacija hiperparametara modela i korištenja različitih tehnika ugrađivanja ulaza. Rezultati najboljeg modela uspoređeni su s rezultatima iz referentnog rada. Arhitektura... More transformera u kombinaciji s podacima nad kojima se koristila k-mer tehnika ugrađivanja ulaza pokazala se jako uspješnom. Less

Abstract (english)

This master thesis focuses on the transformer architecture and the motivation for its use in solving the problem of DNA sequence classification. The architecture was implemented using the Python programming language. The two methods used for the data set preparation are described: the k-mer and the label encoding. The results of variations of model hyperparameters and the use of different input embedding techniques were compared. The best model's results were compared with those from the... More reference work. When combined with data embedded using the k-mer embedding technique, the transformer architecture has proven to be highly effective. Less

• Keywords: Klasifikacija; Bioinformatika; Duboko učenje; Transformer; Ugrađivanje ulaza
• Keywords (english): Classification; Bioinformatics; Deep learning; Transformer; Embedding
• Language: croatian
• URN:NBN: urn:nbn:hr:168:061841
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-10-12 21:30:34