Model dubokog učenja za klasifikaciju MIDI datoteka

master's thesis

• Title (english): A deep learning model for MIDI file classification
• Author: Bartol Rod
• Mentor: Stjepan Šebek (mentor)
• Committee member: Stjepan Šebek (predsjednik povjerenstva)
• Committee member: Stjepan Begušić (član povjerenstva)
• Committee member: Mate Puljiz (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-09-03, Croatia

Abstract

Klasifikacija MIDI datoteki težak je zadatak zbog njihove složenosti. Problem se prvo pokušava riješiti algoritmima strojnog učenja kao što su stroj potpornih vektora, takozvanim plitkim modelima. Zatim se MIDI datoteke tokeniziraju i klasificiraju povratnim neuronskim mrežama. Dana je usporedba oba pristupa i pokazano je kako duboki modeli bolje rješavaju problem klasifikacije MIDI datoteki.

Abstract (english)

Classification of MIDI files is a difficult task due to their complexity. The problem is first attempted to be solved using machine learning algorithms such as support vector machines, so-called shallow models. Then, the MIDI files are tokenized and classified using recurrent neural networks. A comparison of both approaches is given, and it is shown that deep models better solve the problem of classifying MIDI files.

• Keywords: midi; klasifikacija; duboko učenje
• Keywords (english): midi; classification; deep learning
• Language: croatian
• URN:NBN: urn:nbn:hr:168:251682
• 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: Closed access
• Created on: 2024-10-12 22:00:50