Implementacija povratne neuronske mreže i primjena na zadatku predviđanja sljedeće riječi

undergraduate thesis

• Title (english): Implementation of recurrent neural network and application for next word prediction
• Author: Ivan Martinović
• Mentor: Jan Šnajder (mentor)
• Committee member: Jan Šnajder (predsjednik povjerenstva)
• Committee member: Martin Tutek (član povjerenstva)
• Committee member: Josip Jukić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-05, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Duboko učenje danas se intenzivno primjenjuje na različitim zadacima unutar područja umjetne inteligencije. Unutar obrade prirodnog jezika povratne neuronske mreže već duže vrijeme imaju važnu ulogu, stoga je razumijevanje teorije koja stoji iza takvih struktura doista važno. Cilj ovoga rada bilo je upravo dubinsko razumijevanje teorije koja stoji iza povratnih neuronskih mreža te umjetnih neuronskih mreža općenito, a to razumijevanje dobiveno je programskom implementacijom tih modela. Ovaj završni rad podijeljen je u dva dijela. U prvom dijelu predstavljene su slojevita i povratna neuronska mreže te je prikazana njihova implementacija. Drugi dio prikazuje primjenu povratnih neuronskih mreža na zadatku predviđanja sljedeće riječi.

Abstract (english)

Today, deep learning is intensively used for solving different tasks regarding artificial intelligence. Recurrent neural networks have had a significant role in natural language processing for a long time now and that is why understanding the theory behind such structures is of great importance. That is why the main objective of this thesis was to understand the theory behind recurrent neural networks and artificial neural networks in general, which was achieved through the implementation of those models.
The thesis is divided into two major parts. The first part introduces fully connected neural networks and recurrent neural networks, as well as their implementations. The second part focuses on the use of recurrent neural networks in a next word prediction task.

• Keywords: strojno učenje; duboko učenje; povratna neuronska mreža; obrada prirodnog jezika; predviđanje sljedeće riječi; modeli slijed u slijed; generiranje teksta; LSTM
• Keywords (english): machine learning; deep learning; recurrent neural network; natural language processing; next word prediction; sequence-to-sequence models; text generation model; LSTM
• Language: croatian
• URN:NBN: urn:nbn:hr:168:706998
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-14 21:06:31