Automatsko ocjenjivanje testova s ponuđenim odgovorima pomoću dubokog učenja

master's thesis

• Title (english): Automatic Evaluation of Multiple Choice Tests Using Deep Learning
• Author: Ante Grbeša
• Mentor: Marko Subašić (mentor)
• Committee member: Marko Subašić (predsjednik povjerenstva)
• Committee member: Davor Petrinović (član povjerenstva)
• Committee member: Tomislav Pribanić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-07-16, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovome radu razvijena je metoda za automatsko ocjenjivanje testova s ponuđenim odgovorima koja koristi duboko neuronske mreže. Opisane su osnovne značajke neuronskih mreža te konvolucijskih neuronskih mreža. Dan je pregled često korištenih algoritama za detekciju objekata u slikama. Opisan je alat korišten za generiranje korištenog skupa podataka. Opisana je implementacija metode za ocjenjivanje te je dan opis bitnih komponenti metode. Navedene su i opisane metode za evaluaciju metode te rezultati metode na ispitnom skupu.

Abstract (english)

In this paper the method for automatic grading of exams with offered answers was developed using deep convolutional neural networks. Basic characteristics of neural networks and convolutional neural network are described. An overview of commonly used object detection algorithms is given. A tool used for generating data set is described. Also, a description of implementation of method for automatic grading is given along with results for evaluation data set.

• Keywords: neuronske mreže; detekcija objekata; konvolucijske neuronske mreže; računalni vid; TensorFlow Object Detection API; Faster R-CNN; duboko učenje; automatsko ocjenjivanje ispita
• Keywords (english): neural networks; convolutional neural networks; object detection; computer vision; TensorFlow Object Detection API; Faster R-CNN; deep learning; automatic exam grading
• Language: croatian
• URN:NBN: urn:nbn:hr:168:658327
• 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: 2023-01-13 12:20:05