English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Apstrakt: Predavači se svakodnevno suočavaju sa problemom efikasnog prezentovanja materije. Prilikom procene koliko je predavanje u datom trenutku uspešno, predavači se mogu osloniti jedino na svoje iskustvo koje često može odudarati od realnosti. Zbog usredsređenosti na predavanje, predavač nije u mogućnosti da pravovremeno primeti da li je u učionici npr. postalo pretoplo ili zagušljivo, što može nepovoljno da utiče na sposobnost studenata da prate predavanje. Kad bi predavač u svakom trenutku imao objektivniju informaciju o tome da li su studenti zadovoljni, mogao bi je iskoristiti i prilagoditi svoje predavanje i parametre fizičke sredine njihovim potrebama. Ovaj rad ispituje parametre fizičkog okruženja i predavačevog ponašanja koji utiču na zadovoljstvo studenata predavanjem. Izdvojeni su samo oni parametri koji se mogu meriti korišćenjem Interneta inteligentnih uređaja, a analizirani su korišćenjem metoda mašinskog učenja. Cilj je bio da se izdvoje relevantni parametri, a potom da se odabere metoda mašinskog učenja koja daje najbolje rezultate. Ispitano je deset poznatih klasifikatora i njihova sposobnost da na osnovu posmatranih parametara za dati segment predavanja odrede da li pripada klasi u kojoj su studenti zadovoljni predavanjem ili klasi u kojoj nisu zadovoljni. Od ispitanih algoritama, najbolje rezultate imao je Rotation Forest (75.32%), a najslabije Ridor algoritam (70.13%). Abstract: Lecturers regularly face the issue of efficiently presenting some teaching unit. When determining if their presentation is currently successful, lecturers can only rely on their experience which quite often differs from reality. As lecturers are primarily focused on a lecture, they are not able to timely notice if it is too hot or stuffy in the classroom, which may adversely affect the student’s ability to actively attend the lecture. If the lecturer had the more objective information about students’ satisfaction, he/she could use it to adapt the lecture and adjust the parameters of the physical environment to their needs. This paper analyses the parameters of physical environment and lecturer’s behaviour that may affect students’ satisfaction with a lecture. We have selected only those parameters that can be measured by means of Internet of things and analyzed them using the machine learning methods. The aim was to select relevant parameters and choose the most efficient and accurate machine learning method. Ten well-known classifiers were examined and their ability to correctly determine if it belongs to the class in which students are satisfied with the lecture or the other one. Rotation Forest algorithm demonstrated the best (75.32%), while Ridor algorithm (70.13%) the weakest accuracy (70.13%) among the tested classifiers.

Procena zadovoljstva studenata predavanjem korišćenjem metoda mašinskog učenja