Monotonicity in Bayesian Networks for Computerized Adaptive Testing

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F17%3A00313230" target="_blank" >RIV/68407700:21340/17:00313230 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/67985556:_____/17:00476602

Výsledek na webu

<a href="https://link.springer.com/chapter/10.1007/978-3-319-61581-3_12" target="_blank" >https://link.springer.com/chapter/10.1007/978-3-319-61581-3_12</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/978-3-319-61581-3_12" target="_blank" >10.1007/978-3-319-61581-3_12</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Monotonicity in Bayesian Networks for Computerized Adaptive Testing

Popis výsledku v původním jazyce

Artificial intelligence is present in many modern computer science applications. The question of effectively learning parameters of such models even with small data samples is still very active. It turns out that restricting conditional probabilities of a probabilistic model by monotonicity conditions might be useful in certain situations. Moreover, in some cases, the modeled reality requires these conditions to hold. In this article we focus on monotonicity conditions in Bayesian Network models. We present an algorithm for learning model parameters, which satisfy monotonicity conditions, based on gradient descent optimization. We test the proposed method on two data sets. One set is synthetic and the other is formed by real data collected for computerized adaptive testing. We compare obtained results with the isotonic regression EM method by Masegosa et al. which also learns BN model parameters satisfying monotonicity. A comparison is performed also with the standard unrestricted EM algorithm for BN learning. Obtained experimental results in our experiments clearly justify monotonicity restrictions. As a consequence of monotonicity requirements, resulting models better fit data.

Název v anglickém jazyce

Monotonicity in Bayesian Networks for Computerized Adaptive Testing

Popis výsledku anglicky

Artificial intelligence is present in many modern computer science applications. The question of effectively learning parameters of such models even with small data samples is still very active. It turns out that restricting conditional probabilities of a probabilistic model by monotonicity conditions might be useful in certain situations. Moreover, in some cases, the modeled reality requires these conditions to hold. In this article we focus on monotonicity conditions in Bayesian Network models. We present an algorithm for learning model parameters, which satisfy monotonicity conditions, based on gradient descent optimization. We test the proposed method on two data sets. One set is synthetic and the other is formed by real data collected for computerized adaptive testing. We compare obtained results with the isotonic regression EM method by Masegosa et al. which also learns BN model parameters satisfying monotonicity. A comparison is performed also with the standard unrestricted EM algorithm for BN learning. Obtained experimental results in our experiments clearly justify monotonicity restrictions. As a consequence of monotonicity requirements, resulting models better fit data.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/GA16-12010S" target="_blank" >GA16-12010S: Struktury podmíněné nezávislosti: kombinatorické a optimalizační metody</a><br>

Návaznosti

S - Specificky vyzkum na vysokych skolach<br>I - Institucionalni podpora na dlouhodoby koncepcni rozvoj vyzkumne organizace

Rok uplatnění

2017

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Symbolic and Quantitative Approaches to Reasoning with Uncertainty

ISBN

978-3-319-61580-6

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

10

Strana od-do

125-134

Název nakladatele

Springer, Cham

Místo vydání

—

Místo konání akce

Lugano

Datum konání akce

10. 7. 2017

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—