A riveting talk given by Prof. Rajesh Rao on using statistical models to decipher Indus script http://video.ted.com/assets/player/swf/EmbedPlayer.swf

Stumbled on to this piece via 13 Writing Tips- Chuck Palahniuk : Twenty years ago, a friend and I walked around downtown Portland at Christmas. The big department stores: Meier and Frank… Fredrick and Nelson… Nordstroms… their big display windows each held a simple, pretty scene: a mannequin wearing clothes or a perfume bottle sitting in fake snow. But the windows at the J.J. Newberry’s store, damn, they were crammed with dolls and tinsel and spatulas and screwdriver sets and pillows, vacuum cleaners, plastic hangers, gerbils, silk flowers, candy - you get the point.

http://www.vcasmo.com/swf/vcasmo.swf Key Steps : Did not use R for data import operation - Used SPSS to read the data Feature Selection - Used R in this step Data Cleaning - Treatment of Categorical variables was a problem Software used : SAS + R Techniques used : Gradient Boosting machine(gbm package) Rationale : Handling of missing values Robustness against extreme values Handling categorical and continous variables Models interaction between predictors Can model nonlinear dependencies Fitting Time : Couple of hours on a desktop

“ The act of arranging information becomes an act of insight. ” - Edward Tufte

“ When that time comes, I try to be alone and silent for several hours; I need a lot of time to rid my mind of the noise outside and to cleanse my memory of life’s confusion. I light candles to summon the muses and guardian spirits. I place flowers on my desk to intimidate tedium and the complete works of Pablo Neruda beneath the computer with the hope they will inspire me by osmosis.