Sustav za praćenje znanja i preporuku puta učenja

master's thesis

• Title (english): Knowledge Tracing and Learning Path Recommendation System
• Author: Zvonimir Sučić
• Mentor: Goran Delač (mentor)
• Committee member: Goran Delač (predsjednik povjerenstva)
• Committee member: Marin Šilić (član povjerenstva)
• Committee member: Klemo Vladimir (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2022-02-18, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom radu razmatramo problem preporuke puta učenja studenta u okolini e-učenja. Problem dijelimo na potprobleme praćenja znanja i preporuke puta učenja, tako da model za preporuku puta učenja koristi međurezultate modela za praćenje znanja.
Praćenje znanja rješavamo uz pomoć DKVMN, vrste neuronske mreže potpomognute memorijom čija je memorijska struktura zasnovana na listi koncepata znanja iz skupa podataka. Trenirani DKVMN model daje rezultate slične onima u literaturi, što ga čini prihvatljivim za daljnju uporabu. Potproblem preporuke puta učenja rješavamo algoritmom optimizacije strategije TRPO.
Rezultati pokazuju da naš model dugoročno podiže stupanj znanja studenta, i to efikasnije od pohlepnog algoritma odabira zadataka.

Abstract (english)

In this paper we address the problem of learning path recommendation in an e-learning environment. We divide the problem into the problems of knowledge tracing and learning path recommendation, such that our learning path recommendation model uses the results of our knowledge tracing model. We solve the knowledge tracing model by using DKVMN, a special type of a memory augmented neural network whose memory structure is based on a list of knowledge concepts from the dataset. The trained DKVMN model gives results similar to ones in the literature, making it adequate for further use. We solve the problem of learning path recommendation with the policy optimization algorithm TRPO. The results show that our model will, in the long term, increase the knowledge levels of a student, and do so more efficiently than the greedy baseline algorithm of problem selection.

• Keywords: sustav preporuka; praćenje znanja; preporuka puta učenja; DKVMN; TRPO
• Keywords (english): recommender system; knowledge tracing; learning path recommendation; DKVMN; TRPO
• Language: croatian
• URN:NBN: urn:nbn:hr:168:025361
• 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 13:51:04