Automatizirano vrednovanje dijagnostičkih vještina zasnovano na ispitanikovom oblikovanju pravila za rješavanje scenarijskih problemskih zadataka

doctoral thesis

• Title (english): Automated diagnostic skill assessment based on examinee's explication of rules for scenario-based problem solving tasks
• Author: Juraj Petrović
• Mentor: Predrag Pale (mentor)
• Committee member: Predrag Pale (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronic Systems and Information Processing); Zagreb
• Defense date and country: 2017, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing; Data Processing
• Universal decimal classification (UDC): 621.3 - Electrical engineering

Abstract

Cilj je ovoga doktorskog istraživanja bio unaprijediti stanje znanosti u području računalno potpomognute provjere znanja razvojem alata koji bi omogućio automatiziranu samoprovjeru ispitanikovih dijagnostičkih vještina i pružanje odgovarajuće povratne informacije. Da bi se provjerila važnost i opravdanost razvoja takvog alata, provedena su dva podistraživanja u kojima je najprije ustanovljena potreba za alatom za samoprovjeru znanja tijekom njegova usvajanja, a potom je ispitan potencijal računalnih samoprovjera znanja. Temeljem rezultata podistraživanja potvrđena je važnost alata koji bi omogućio domenski neovisnu provjeru dijagnostičkih vještina uz automatsko generiranje povratne informacije. Alat koji se u ovoj disertaciji predlaže, modelira i vrednuje, temelji se na metodi automatiziranog vrednovanja pravila za rješavanje scenarijskih problemskih zadataka. Pravila, koja ispitanik definira u obliku stabla odluke na temelju ključnih parametara i vrijednosti koje oni mogu poprimiti, pritom se vrednuju u odnosu na dostupno referentno znanje. Prototipna implementacija alata koji podržava ovu metodu provjere vrednovana je na skupini ispitanika. Rezultati vrednovanja pružili su podršku za tvrdnje da je modelirani način provjere vremenski i radno efikasan, budući da ispitanik ne rješava samo jedan primjer problema nego demonstrira način na koji općenito donosi zaključak u skupini srodnih dijagnostičkih problema. Također, rezultati vrednovanja pokazali su da ovakav način provjere ispitanicima ne uzrokuje preveliko kognitivno opterećenje i da im može pružiti razumljivu i relevantnu povratnu informaciju.

Abstract (english)

The aim of this doctoral research was to improve the state of knowledge in the field of computer-based knowledge assessment, specifically formative self-assessments, by modelling and developing a tool for self-assessment of diagnostic skills. Two initial studies were performed to verify the importance and justification for the development of a tool for self-assessment of diagnostic skills with automated feedback generation. The first one, in order to establish the existence and importance of a need for knowledge self-assessment during its acquisition, and the second one, in order to examine the potential of knowledge self-assessment in achieving the desired learning outcomes. The results obtained in both studies justified the development and modelling of the diagnostic skills self-assessment method proposed in this dissertation. The diagnostic skills self-assessment method proposed in this dissertation is based on defining the set of rules for solving scenario-based diagnostic tasks, depending on the key properties or tests related to the diagnostic problem, and the exact values of those properties or tests. The rules, defined in form of a decision tree, are then automatically evaluated with regard to the referent expert knowledge stored implicitly in available diagnostic test cases. The prototype system implementation and its properties were evaluated on a group of subjects who had to solve a diagnostic problem using the proposed method and have their solution evaluated by the system. The results of the evaluation provided support for the claim that the proposed way of self-assessment was time efficient, does not cause too much cognitive load, and that it can provide its users with automatically generated, comprehensive and relevant feedback.

• Keywords: računalno potpomognuta provjera znanja; formativna provjera znanja; samoprovjera; provjera dijagnostičkih vještina; stabla odluke; rješavanje problema
• Keywords (english): computer-based knowledge assessment; formative assessment; knowledge self-assessment; diagnostic skills assessment; decision trees problem solving
• Language: croatian
• URN:NBN: urn:nbn:hr:168:703974
• Study programme: Title: Electrical Engineering and Computing; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti elektrotehnike i računarstva (Doktor znanosti elektrotehnike i računarstva)
• Type of resource: Text
• Extent: 110 str. ; 30 cm.
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-04-09 15:23:58