Causal network knowledge

Component identification and knowledge. Assembly-statistically significant components differing between control and perturbed cohorts are identified. Ultimately correlation or causal networks are interro-... 

The classification of human error within the FASGEP project takes into account the woric of Rasmussen, specifically the classification of Rule-Based errors, Skill-Based errors, and Knowledge-Based errors. The causal relationships for each of the classifications have been developed into a causal network. A type of graphical probability model is based on this. 

The proposed model consists of two components a system d3mamics framework and a Bayesian belief network structure. The system dynamics formalism enables us to represent change over time and change due to feedback. The Bayesian behef network formalism enables us to represent networks of causality and capture stocasticity and uncertainty. BBNs also enable us to incorporate new knowledge and update the model as new evidence becomes available. Combining SD and BBN has been previously proposed and applied in aviation risk context (Mohaghegb, et.al., 2008). 

From these definitions we can see that an indispensable component to both early warning and proactive control systems is the set of causal relations between deviations in food supply networks and determinant factors. Ultimately, expert domain knowledge is required to verify those causal relations. However, current advances of ICT systems provide the opportunities to largely automatically disentangle causal relations from huge amounts of data. 

A Bayesian Network (BN) (E. Zio 2009) as a probability-based knowledge representation method is appropriate for the modeling of causal processes with uncertainty. The BN is composed of nodes (variables) and arcs linking the nodes. It is a Directed Acyclic Graph (DAG) whose nodes represent random variables and links define probabilistic dependencies between variables. When the node obtain the available data (or evidence), the instantiation of that node occurs. Note that the amount or quality of the data is not required. That means the available data in this research is assumed as representative of the phenomenon observed and what is being modeled. By the Ref (Jensen, Finn V., 2001), the heart of Bayesian network is building... 

Li J, Shi J., 2007. Knowledge discovery from observational data for process control using causal Bayesian networks. HE Trans, 39(6) 681-690. 

---