A New Score for Adaptive Tests in Bayesian and Credal Networks

  title={A New Score for Adaptive Tests in Bayesian and Credal Networks},
  author={Alessandro Antonucci and Francesca Mangili and Claudio Bonesana and Giorgia Adorni},
A test is adaptive when its sequence and number of questions is dynamically tuned on the basis of the estimated skills of the taker. Graphical models, such as Bayesian networks, are used for adaptive tests as they allow to model the uncertainty about the questions and the skills in an explainable fashion, especially when coping with multiple skills. A better elicitation of the uncertainty in the question/skills relations can be achieved by interval probabilities. This turns the model into a… 
Intelligent Tutoring Systems by Bayesian Nets with Noisy Gates
This work advocates logical gates with uncertainty for a compact parametrization of the conditional probability tables in the underlying Bayesian net used by tutoring systems and derives a dedicated inference scheme to speed up computations.
ADAPQUEST: A Software for Web-Based Adaptive Questionnaires based on Bayesian Networks
ADAPQUEST, a software tool written in Java for the development of adaptive questionnaires based on Bayesian networks, is introduced and an application of this tool for the diagnosis of mental disorders is discussed.


Reliable Knowledge-Based Adaptive Tests by Credal Networks
This work suggests the use of credal networks, a generalization of Bayesian networks based on sets of probability mass functions, to implement adaptive tests exploiting the knowledge of the test developer instead of training on databases of answers.
Monotonicity in practice of adaptive testing
This article evaluates Bayesian network models used for computerized adaptive testing and learned with a recently proposed monotonicity gradient algorithm, which has a lower question prediction quality than unrestricted models but is better in the main target, which is the student score prediction.
Exploiting Bayesian Network Sensitivity Functions for Inference in Credal Networks
A preprocessing step is proposed that is able to reduce the complexity of Bayesian network computations and shows that for some classes of parameter in Bayesian networks the qualitative effect of a parameter change on an outcome probability of interest is independent of the exact numerical specification.
Modeling Unreliable Observations in Bayesian Networks by Credal Networks
This work proposes a procedure for a more general modeling of the observations, which allows for updating beliefs in different situations, including various cases of unreliable, incomplete, uncertain and also missing observations, and transforms the original Bayesian network into a credal network.
Probabilistic Inference in Credal Networks: New Complexity Results
It is shown that inferences under strong independence are NP-hard even in trees with binary variables except for a single ternary one, and it is proved that under epistemic irrelevance the polynomial-time complexity of inferences in credal trees is not likely to extend to more general models.
Computer Adaptive Testing Using the Same-Decision Probability
It is shown empirically that utilizing the Same-Decision Probability is a viable and intuitive approach for determining question selection in Bayesian-based Computer Adaptive Tests, as its usage allows us to ask fewer questions while still maintaining the same level of precision and recall in terms of classifying competent students.
How To Grade a Test Without Knowing the Answers - A Bayesian Graphical Model for Adaptive Crowdsourcing and Aptitude Testing
An active learning/adaptive testing scheme based on a greedy minimization of expected model entropy is devised, which allows a more efficient resource allocation by dynamically choosing the next question to be asked based on the previous responses.
Approximating Credal Network Inferences by Linear Programming
Preliminary empirical results suggest that the accuracy of the optimal class set is seldom affected by the approximate probabilities, and the approach can also be specialized to classification with credal networks based on the maximality criterion.
Bayesian Networks In Educational Testing
  • J. Vomlel
  • Computer Science
    Int. J. Uncertain. Fuzziness Knowl. Based Syst.
  • 2002
The experiments suggest that the test design can benefit from a Bayesian network that models relations between skills, not only in the case of an adaptive test but also when designing a fixed (non-adaptive) test.