About Bayesian statistics

Over the last 25 years, Bayesian networks have emerged as a practically feasible form of knowledge representation. With the ever-increasing computing power, Bayesian networks are now a powerful tool enabling deep understanding of complex, multi-dimensional problem domains (Lynch., 2007). Their computational efficiency and inherently visual structure make Bayesian networks attractive for exploring and explaining complex problems.

The proposed models or systems in the present services are characterized by a multi-level hierarchical structure that embraces a large number of components interacting with each other and jointly contributing to the functionality of a specific subsystem, and, contributing to the functionality of the whole system, as applied to ecology ( S. Clark, 2005).

The key advantage of probabilistic approaches like Bayesian is that they can take a large body of data; Bayesian paradigm could deduce likely relationships on its own as opposed to “frequentist”, where each causal connection has to be specifically taught. The model could be well suited for capturing probabilistic and incomplete causal knowledge regarding a Public Health Programme. Bayesians finding computational techniques that now allow analysts the opportunity to connect the methods with the demands of practical science.

Overall, a multi-level hierarchical structure is typical for systems that are designed to meet certain highly complex and advanced functional demands, such as the systems in aerospace or defense industries (troubleshooting, study of defects, process optimization, risk analysis, etc.), health (biochip analysis, characterization of illnesses and treatments, etc.), marketing (driver analysis, scoring, customer and product segmentation, etc.) and risk management and many other domains of activities.

A few application examples with Bayesian Networks in health sector:

  • Cancer classification by means of microarray analysis: Microarray analysis is a technique for gene expression profiling of cell samples, expression profiles indicate which genes are currently active among thousands of genes.
  • Breast Cancer Diagnostics (through Wisconsin Breast Cancer Database)
  • Difficult intubations analysis. Prediction of difficult intubation (DI) is crucial during pre an aesthesia assessment of a patient. Many criterions are used to predict DI with different performances. BayesiaLab, through its learning algorithms, allows to quickly discovering unknown probabilistic relationships between variables and enhance prediction.
  • Salmonella isolation. Identification of factors associated with Salmonella isolation on pork carcasses.
  • Biocomputing transcriptome analysis Bioinformatics.

Laisser un commentaire

Votre adresse de messagerie ne sera pas publiée. Les champs obligatoires sont indiqués avec *

Vous pouvez utiliser ces balises et attributs HTML : <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Post Navigation