China
Africa
Explore chapters and articles related to this topic
AI-Based Predictive Models for Adaptive Learning Systems
Published in Prathamesh Churi, Shubham Joshi, Mohamed Elhoseny, Amina Omrane, Artificial Intelligence in Higher Education, 2023
Prashant Gupta, Trishul Kulkarni, Bhagwan Toksha
Schiffaino et al. developed an intelligent teaching agent called eTeacher, built to assist e-learners in a personalized manner and tested for Systems Engineering course (Schiaffino et al., 2008). The learner profile was built with the help of observations in a learner’s behaviour while they take online classes. The profile includes learning style, performance assessment for exercises done, topics undertaken for study, results of exam conducted, and so on. Learning style is automatically perceived through actions using Bayesian networks which, along with other information in the learner profile, assists the eTeacher to engage and help the learner on a proactive basis by suggesting them personalized courses that will assist them during the process of learning. The interdependencies between the learning styles and behaviours are encoded in the Bayesian model as shown in Figure 6.6 via the arcs that travel from the nodes indicative of learner behaviour to the nodes indicative of learning style dimensions. It is also indicative of a conditional probability table for the understanding node.
Study on system testing approach-based risk evaluation of subsea tree
Published in Chongfu Huang, Zoe Nivolianitou, Risk Analysis Based on Data and Crisis Response Beyond Knowledge, 2019
Xiaobing Yuan, Baoping Cai, Guoming Chen
The graphs of the network nodes represent the variables, and the arrows with directions indicate the dependencies and causal relationships between the nodes. This quantitative relationship is established by the conditional probability table. In general, the conditional probability table is determined by historical data and expert judgment. At the same time, once the corresponding nodes change, the nodes associated with them also change accordingly, and the real-time status of each node can be changed by the influence of each other.
Branching rules and quantification based on human behavior in the ADS-IDAC dynamic PRA platform
Published in Stein Haugen, Anne Barros, Coen van Gulijk, Trond Kongsvik, Jan Erik Vinnem, Safety and Reliability – Safe Societies in a Changing World, 2018
To reduce the conditional probability table size, the noisy OR gates can be used to specify the DBN and to build the conditional probability table for the crew failure modes nodes. In relation to the DBN quantification model included in ADS-IDAC, the leaky noisy OR gate also give the advantage of representing the probability that a crew failure can occur even when there is no influence from any of the PSFs. In other words, the leak factor provides a way to include other PSFs that are not explicitly represented in the DBN model as individual PSF nodes.
Influence of building parameters on energy efficiency levels: a Bayesian network study
Published in Advances in Building Energy Research, 2022
Lakmini Rangana Senarathne, Gaurav Nanda, Raji Sundararajan
Each node of the Bayesian network graph has consisted of a conditional probability table, which expresses the relationship between nodes. The purpose of the conditional probability table is to define the behaviour of each variable with its parents. To generate the probability table, the total range of each variable has been split into smaller sections. Table 2 prototypes the split sections of variables according to their total ranges. Only Xc and Xe variables are considered, and it is assumed that their ranges are spanning between [Xc,L, Xc,H] and [Xe,L, Xe,H], where XL is the minimum and XH is the maximum value of each variable. Xe and Xc variables are divided into the number of m and n sections.