Methoden vergelijken

	Bayesiaans Netwerk	LSTM	Rasch Model
Vakgebied	Bayesiaanse statistiek	Deep learning	Psychometrie
Familie	Bayesian methods	Machine learning	Latent structure
Jaar van ontstaan	1988	1997	1960
Grondlegger	Judea Pearl	Hochreiter, S. & Schmidhuber, J.	Georg Rasch
Type	Probabilistic graphical model	Recurrent neural network (gated memory cell)	Item Response Theory / Latent trait model
Oorspronkelijke bron	Pearl, J. (1988). Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference. Morgan Kaufmann. ISBN: 978-1558604797	Hochreiter, S. & Schmidhuber, J. (1997). Long Short-Term Memory. Neural Computation, 9(8), 1735–1780.	Rasch, G. (1960). Probabilistic Models for Some Intelligence and Attainment Tests. Danish Institute for Educational Research, Copenhagen.
Aliassen	Bayes network, belief network, probabilistic graphical model, directed graphical model	LSTM (Uzun Kısa Dönem Bellek Ağı), long short-term memory, LSTM network, recurrent neural network with memory cells	1PL IRT, one-parameter logistic model, Rasch Modeli — 1PL IRT, 1PL model
Verwant	4	5	6
Samenvatting	A Bayesian network is a probabilistic graphical model, introduced by Judea Pearl in 1988, that encodes a set of variables and their conditional dependencies as a directed acyclic graph (DAG). Each node represents a variable; each directed edge encodes a direct probabilistic influence. By combining Bayes' rule with the graph's conditional independence structure, the model supports reasoning under uncertainty — computing the probability of any variable given observed evidence about others.	LSTM (Long Short-Term Memory) is a recurrent neural network architecture, introduced by Sepp Hochreiter and Jürgen Schmidhuber in 1997, that can learn long-term dependencies in sequential data and is widely used for time-series and sequence prediction. It keeps an internal memory that lets information persist across many time steps.	The Rasch model, introduced by Georg Rasch in 1960, is the simplest member of the Item Response Theory (IRT) family. It assigns a single difficulty parameter to each test item and places both item difficulties and person abilities on the same logit scale, enabling direct, sample-independent comparison of items and persons.