Home  ›  Resource  ›  Lectures  ›  Content

Applications of Bayesian Networks to Small, Mid-size and Massive Data and an introduction to FuturICT

Title: Applications of Bayesian Networks to Small, Mid-size and Massive Data and an introduction to FuturICT

Speaker:  Ron S. Kenett

Time:04/24/2012 10:00am-12:00am

Location: FIT 4-302

Abstract:Modelling cause and effect relationships has been a major challenge for statisticians in a wide range of application areas. Bayesian Networks (BN), Context Statistical Process Control (CSPC),Variable Order Bayesian Networks (VOBN) and Target-Based Bayesian Networks combine graphical analysis with Bayesian analysis to represent causality maps linking measured and target variables (Pearl, 2000, Ben Gal, 2003, 2004, 2007, Kenett, 2007, Cornalba et al, 2007, Gruber and Ben-Gal 2011). Such mapping models can be used for effective process monitoring. The talk will present an introduction to Bayesian Networks and their applications in the context of the amount of data available. We distinguish between small data sets, with multivariate observations over 5-15 time instances, mid-range data sets with data collected over 100-500 time stamps and massive data sets with thousand and even millions data points. The examples we cover include the testing of web services (Bai and Kenett, 2012), bioinformatics (Grau et al, 2008) and web site usability analysis with massive data sets (Harel et al, Kenett et al, 2009), customer satisfaction surveys (Kenett and Salini, 2011) and operational risks with midrange data sets (Kenett and Raanan, 2010), healthcare systems (Kenett, 2012), and biotechnology with small data sets (Peterson and Kenett, 2011). We will discuss various challenges in dealing with complex data sets and provide an introduction to the European ambitious FuturICT project (www.futurict.eu) that aims at handling planet level data sets.

Bio:Ron Kenett,Chairman and CEO of the KPA Group, Research Professor, University of Torino, Turin, Italy, International Professor Associate, Center for Risk Engineering, NYU-Poly, NY, USA and President of the Israel Statistical Association (ISA). Ron is editor in chief of the Wiley Encyclopedia of Statistics in Quality and Reliability, member of the advisory editorial board of the international journal Quality Technology and Quantitative Management, associate editor of the ISI ISBIS journal Applied Stochastic Models in Business and Industry (ASMBI), Past President of the European Network for Business and Industrial Statistics (ENBIS) and member of the board of directors of Optimata and the publicly traded biotech company D-Pharm.