The LESO-PB building building control system 0.0015 Density - PowerPoint PPT Presentation

Integrating R in an advanced The LESO-PB building building control system 0.0015 Density estimate 0.0010 0.0005 0.0000 0 500 1000 1500 2000 2500 3000 3500 [lux] David Lindelöf David Lindelöf useR! 2006 useR! 2006 15/06/2006 Vienna 15/06/2006 Vienna david.lindelof@epfl.ch david.lindelof@epfl.ch Bayes’s theorem User discomfort probability = = = Pr( ) C False E e = = = Pr( ) Pr( ) 1.0 E e C False C False 0.8 = = = + = = = Pr( ) Pr( ) Pr( ) Pr( ) Discomfort probability E e C False C False E e C True C True 0.6 0.4 Illuminance distribution for Illuminance distribution for 0.2 uncomfortable situations comfortable situations 0.0 0 500 1000 1500 2000 2500 3000 3500 [lux] cf. email classifiers: Pr( = = " " ) Spam True Content Nigeria David Lindelöf David Lindelöf useR! 2006 useR! 2006 15/06/2006 Vienna 15/06/2006 Vienna david.lindelof@epfl.ch david.lindelof@epfl.ch

Density estimation with Davies & Density estimation methods Kovac’s ftnonpar package True density Spline SHK97 Taut string 0.004 12 12 12 0.003 10 10 10 Density estimate = = Pr( ) 8 8 8 E e C False 0.002 6 6 6 4 4 4 0.001 2 2 2 0.000 0 0 0 0 500 1000 1500 2000 2500 3000 3500 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 [lux] Histogram Kernel SJ91 Total variation 12 12 12 0.0015 10 10 10 Density estimate 8 8 8 = = 0.0010 Pr( ) E e C True 6 6 6 4 4 4 0.0005 2 2 2 0.0000 0 0 0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0 500 1000 1500 2000 2500 3000 3500 [lux] David Lindelöf David Lindelöf useR! 2006 useR! 2006 15/06/2006 Vienna 15/06/2006 Vienna david.lindelof@epfl.ch david.lindelof@epfl.ch Discomfort probability Combination of variables 1.0 1500 0.8 = = Discomfort probability Pr( ) C False E e Eye−level illuminance [lux] 1000 Horizontal and eye-level 0.6 illuminances 0.4 500 0.2 0 0.0 0 500 1000 1500 0 500 1000 1500 2000 2500 3000 3500 Horizontal workplane illuminance [lux] [lux] David Lindelöf David Lindelöf useR! 2006 useR! 2006 15/06/2006 Vienna 15/06/2006 Vienna david.lindelof@epfl.ch david.lindelof@epfl.ch

Summary ● R called from a Java program to perform density estimation with taut-string algorithm ● Data prepared with an R script, then Davies & Kovac’s ftnonpar package computes the estimation David Lindelöf useR! 2006 15/06/2006 Vienna david.lindelof@epfl.ch