Using Bayesian Methods for Predicting the Development of Children Autism

Mariia A. Voronenko, Iryna Lurie, Oleg O. Boskin, Ulzhalgas Zhunissova, R.V. Baranenko, Volodymyr I. Lytvynenko

2019

Closed access Metadata (CSL-JSON)
DOI 10.1109/atit49449.2019.9030523
Data confidence 70 medium
How this score is calculated

It is built from checkable properties of the record itself: whether the article has a DOI, how many independent sources confirmed its data, whether the required fields are filled in, and how recent the latest confirmation is. Disagreement between sources and a missing identifier lower it. A value of 80 or above means the record was published without further checking; 60 to 79 means it is published at reduced confidence, with a manual check advisable. A record verified by an editor by hand is never pushed below 90 by a recalculation. This scores our data about the article, not the article itself.

Abstract

In this paper, we propose a methodology for using static Bayesian networks (BN) in the tasks of predicting the development of the children's disease “autism” aged 4 to 11 years. Methods of constructing the structure of a static BN, parametric learning, validation are considered, a sensitivity analysis and scenario analysis of “What-if' are carried out. The model was designed in collaboration with pediatricians, as well as subject matter experts in the selection and quantification of input and output variables.