Predicting system collapse: Two theoretical models

Abstract

The intricate construction of online educational systems lies within three principal activities: Design , implementation and proper post-implementation assessment. There is not enough knowledge or experience in those regards. Efficient execution of these three major activities necessitates the use of design and pedagogical models to achieve cost and time efficiency, as well as high pedagogical quality. Utilization of online educational systems would benefit from a struc-tured approach to design, implementation, and student's assessment. In this paper, we present the design of an online education system and its implementation, and we analyze student's behavior towards the system using the theory of planned behavior and the technology acceptance model. A survey methodology approach was followed. The partial least squares method was used for the assessment of the results discussed. Structural equation analysis provides evidence for the superiority of the theory of planned behavior in explaining student's behavior towards online educational systems. Limitation, implications, design recommendations, and suggestions for future research are discussed.

Characterization of Inclination Analysis for Predicting Onset of Heart Failure from Primary Care Electronic Medical Records

“…In this section, a brief outline of inclination analysis is reported. A detailed description can be found in [12,13]. An exemplary representation of inclination analysis is shown in Figure 1.…”

“…In this first exploratory study, the parameter δ was empirically set equal to 0.1 to account for the inherent uncertainty of the clinical data, whereas in the original formulation, δ was set equal to 0 [12,13]. The algorithm was implemented in MATLAB R2020a.…”

“…The inclination analysis was first designed by Kryazhimskii and Beck in 2002 [12] to model the behaviour of environmental systems (e.g., survival/collapse). The same approach was later applied to analyze financial time series [13], but to the best of our knowledge, so far, the inclination analysis has not been applied to develop health prediction models. An inclination analysis is based on a time series path-dependent stochastic process theory to predict the future behaviour of a system from past observations.…”

“…(iii) Clinical data are multivariate in nature as multiple biomarkers are monitored in time, whereas in the original formulation of the inclination analysis, univariate time series are analysed (e.g., rodent population size in environmental behaviour prediction, exchange rates and closing stock prices in financial prediction) [12,13]. Therefore, to apply an inclination analysis to multivariate clinical data, novel approaches should be defined (e.g., classification models able to deal with multiple features per record).…”