BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Alternative Latent and Observable Factors for Knowledge Tracing\; 
 A time series approach
DTSTART:20210713T100000
DTEND:20210713T120000
DTSTAMP:20260603T161447Z
UID:924801e5dde8c28c478f32879c2c2777196fde33dcf4251008c0e5a8
CATEGORIES:Conferences - Seminars
DESCRIPTION:Jade Cock\nEDIC candidacy exam\nexam president: Prof. Martin J
 aggi\nthesis advisor: Prof. Tanja Käser\nco-examiner: Prof. Pierre Dillen
 bourg\n\nAbstract\nInteractive simulations can foster inquiry learning\,\n
 but the complexity of those environments require adaptive\nguidance for th
 e students to efficiently use those systems. The\ngoal of our thesis is to
  build such an adaptive platform. To guide\nus in our future endeavours\, 
 we examine three papers. The first\none is a review providing guidance abo
 ut methodological choices\nin the context of pipelines involving learner m
 odels. The second\none is a working example of the application of a cluste
 r-rule\nmining-classification framework. The last one is a novel approach\
 nto classify asynchronous and irregular time series. We end with\nour thes
 is proposal.\n\nBackground papers\n\n	Pelánek\, R. (2017). Bayesian knowl
 edge tracing\, logistic models\, and beyond: an overview of learner modeli
 ng techniques. User Modeling and User-Adapted Interaction\, 27(3)\, 313-35
 0.\n	Horn\, M.\, Moor\, M.\, Bock\, C.\, Rieck\, B.\, & Borgwardt\, K. (20
 20\, November). Set functions for time series. In International Conference
  on Machine Learning (pp. 4353-4363). PMLR.\n	Fratamico\, L.\, Conati\, C.
 \, Kardan\, S.\, & Roll\, I. (2017). Applying a framework for student mode
 ling in exploratory learning environments: Comparing data representation g
 ranularity to handle environment complexity. International Journal of Arti
 ficial Intelligence in Education\, 27(2)\, 320-352.\n
LOCATION:
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR