Model prilagodljivoga stjecanja znanja učenika u sustavima e-učenja

doctoral thesis

• Title (english): Adaprive student's knowledge acquisition model in e-learning systems
• Author: Ani Grubišić
• Mentor: Slavomir Stankov (mentor)
• Committee member: Vlado Glavinić (predsjednik povjerenstva)
• Committee member: Leo Budin (član povjerenstva)
• Committee member: Robert Manger (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronics, Microelectronics, Computer and Intelligent Systems); Zagreb
• Granter: University of Split; Faculty of Science; Split
• Defense date and country: 2012-02-24, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

Model automatskog i dinamičkog generiranja prilagodljivog računalom oblikovanog nastavnog sadržaja u sustavima e-učenja, AC-ware Tutor, predstavlja sinergiju tradicionalne strukture inteligentnih tutorskih sustava, stereotipova, Bloomove taksonomije znanja i Bayesovih mreža. U današnje vrijeme, kada se najviše uočava problematika prilagođavanja znanju učenika, ovakav model donosi rješenja koja olakšavaju učiteljima dugotrajan i iscrpljujući proces oblikovanja nastavnih sadržaja i testova. Naime, automatizam u generiranju elemenata računalom oblikovanog nastavnog sadržaja prebacuje spomenuti zadatak s živih učitelja na računalnog tutora, koji samostalno oblikuje, odabire i niže elemente računalom oblikovanog nastavnog sadržaja potpuno prilagođene razini znanja učenika. Razina znanja učenika je određena njegovim stereotipom, a profinjena je Bayesovim modelom učenika. Ovakav računalom oblikovani nastavni sadržaj nije statičan, već se dinamički mijenja nakon svakog provjeravanja znanja, prilagođavajući sljedeći element courseware-a za učenje aktualnoj razini znanja učenika. Učenik se uči i poučava onoliko dugo koliko mu je potrebno da postigne željenu razinu znanja. Poželjno je da svaki učenik završi cijeli proces učenja i poučavanja na razini stručnjaka, ali kako to nije ostvarivo ni u okruženju tradicionalne nastave, tako se i u ovom pristupu omogućava učeniku završavanje procesa učenja i poučavanja na nižim razinama znanja. Još jednom naglašavamo da model AC-ware Tutor prilagođava sadržaj, a ne sučelje, znanju učenika, a ne stilovima učenja, što ga čini jedinstvenim primjerkom među prilagodljivim sustavima e-učenja.

Abstract (english)

A model of automatic and dynamic generation of adaptive courseware in e-learning systems, AC-ware Tutor, represents a synergy between traditional intelligent tutoring systems structure, stereotypes, Bloom’s knowledge taxonomy and Bayesian networks. Nowadays, when student’s knowledge adaptation problematic emerges the most, this model brings solutions to teachers easing long and exhausting process of designing courseware and tests. Namely, automatism in generation of courseware elements is, in this model, transferred from teachers to computer tutor, who designs, selects and sequences courseware elements adapted to student’s knowledge level. Student’s knowledge level is determined by his or her stereotype, and refined by Bayesian student model. Courseware designed in this model is not static, but it changes dynamically after every knowledge test, adapting next courseware element to current student’s knowledge level. Student is involved in learning and teaching process as long as it is necessary for him or her to acquire wanted knowledge level. It is desirable that each and every student finishes learning and teaching process as expert, but as it is not achievable in traditional classroom, this model enables student to finish learning and teaching process at lower knowledge levels. We emphasise once more that AC-ware Tutor model adapts content, not interface, to student’s knowledge, not learning styles, what makes it unique exemplar among adaptive e-learning systems.

• Keywords: inteligentni tutorski sustav; prilagodljivi courseware; automatsko generiranje pitanja; Bloom-ova taksonomija znanja; Bayes-ova mreža predikcija; stereotip
• Keywords (english): intelligent tutoring systems; adaptive courseware; automatic question generation; Bloom's knowledge taxonomy; Bayesian probability network; stereotype
• Language: croatian
• URN:NBN: urn:nbn:hr:168:865834
• Project: Number: 177-0361994-1996; Title: Oblikovanje i vrednovanje inteligentnih sustava e-učenja.; Leader: Branko Žitko; Jurisdiction: Croatia; Funder: MZOS; Funding stream: ZP
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Catalog URL: http://lib.fer.hr/cgi-bin/koha/opac-detail.pl?biblionumber=38848
• Type of resource: Text
• Extent: 279 str ; 30 cm
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-07-17 07:47:12