Uporaba pametnih tehnologij pri poučevanju ter pri analizi učnega procesa
Izobraževanje je v moderni družbi povsod prisotna aktivnost, ki se izvaja tekom celega življenja. Psihološka in pedagoška znanost se intenzivno ukvarjata z izobraževalnim procesom, toda pri tem manjkajo realne meritve tako posameznikov – učencev, kot njihove skupinske dinamike znotraj učnega procesa.
Naše raziskovalno vprašanje je torej, kako izdelati učinkovit sistem merjenja parametrov posameznika znotraj učnega procesa, kako analizirati meritve ter razviti modele, ki bodo napovedovali pomembne parametre, kot je na primer posameznikova pozornost.
V ta namen smo razvili sistem snemanja in analize obnašanja učencev (telo, pogled) z uporabo Kinect senzorja. Na osnovi zajetih podatkov pogleda in gibanja telesa smo izdelali prediktivni model napovedovanja pozornosti učenca. Analizirali smo pozornost učencev med poukom ter raziskali, ali ta vpliva na učni izid oziroma uspešnost poučevanja.
Drugi pomembni izziv pa je uporaba sodobnih pametnih tehnologij znotraj učnega procesa, ki naj tudi motivirajo učenca in povečajo njegovo sodelovanje. S tem ciljem smo razvili sistem Typing tutor, ki motivira pri treningu tipkanja s pomočjo socialne komunikacije na principu pozitivnega povratnega odziva.
Sodelavci
Odprta vprašanja in raziskovalni rezultati
Raziskovalni rezultati in članki
Weak Ground Truth Determination of Continuous Human-Rated Data
A research paper (10.1109/ACCESS.2020.3046293) and web tool (https://www.lucami.org/en/WGT/) on Weak Ground Truth Determination of Continuous Human-Rated Data. The article presents a novel weak ground truth (WGT) determination and rater bias removal procedure on continuous human-rated data. The notion of WGT is essential in cases where there is no direct empirical…
Research paper on predicting students’ attention
Research paper Predicting students’ attention in the classroom from Kinect facial and body features was published in the EURASIP Journal on Image and Video Processing, vol. 80, 2017. Abstract This paper proposes a novel approach to automatic estimation of attention of students during lectures in theclassroom. The approach uses 2D…
Paper on student project building up student’s career prospects
Urban Burnik, Andrej Košir: Industrial product design project: building up engineering students’ career prospects The needs to bridge the academic skills of students involved in higher education and improve their readiness for employment in industry are being widely addressed. This paper presents results of a specific engineering project case supervised by…
Paper on emotion elicitation in socially intelligent services
Andrej Košir, Gregor Strle: Emotion Elicitation in a Socially Intelligent Service: The Typing Tutor The aim of the study is to evaluate the extent to which the machine emotion elicitation can influence the affective state (valence and arousal) of the learner during a tutoring session. The tutor provides continuous real-time…
Paper on student attention related to their learning gain
Urban Burnik, Janez Zaletelj, Andrej Košir: Video-based learners’ observed attention estimates for lecture learning gain evaluation This paper suggests an objective and non-intrusive evaluation of learners’ attention against learning outcomes by introducing an observed attention estimate (OAE). The procedure uses human annotations based on visual cues with a supporting video…
Tema: Razvoj modela pozornosti študentov
Problem: Na podlagi video posnetkov študentov med lekcijo v razredu je potrebno izdelati model, ki napoveduje njihovo pozornost (oziroma engagement – angažiranost). Upoštevati je potrebno različne vizualne znake (aktivnosti, kot so pisanje, poslušanje), znake telesa (body language), mimiko obraza, ter znake distrakcij (motnje, utrujenost..). Podatki: Obstajajo video posnetki 18 študentov…
Podatkovni set Študenti v razredu (Kinect)
Podatkovni set zajema video posnetke 6 študentov med učno uro v razredu. Za vsakega so dodatno podane točke skeleta (telesa), ki jih je detektiral senzor Kinect, potem obrazne točke, ter dodatni signali smeri pogleda ipd. Opis podatkovnega seta: