Uma aplicação interativa em Shiny, R e Radarchart para visualização de Diagramas Kiviat

Na publicação anterior (Diagramas Kiviat (Spider Charts) em R!)  vimos como é simples produzir  um diagrama de Kiviat ou Radar Chart no R com o pacote radarchart.

continuar lendo   >>>>

The Improbability Principle: Why Coincidences, Miracles, and Rare Events Happen Every Day, David J. Hand –

June 27, 2017 no comments Posted in Análise de Dados

In The Improbability Principle, the renowned statistician David J. Hand argues that extraordinarily rare events are anything but. In fact, they’re commonplace. Not only that, we should all expect to experience a miracle roughly once every month.      But Hand is no believer in superstitions, prophecies, or the paranormal. His definition of “miracle” is thoroughly rational. No mystical or supernatural explanation is necessary to understand why someone is lucky enough to win the lottery twice, or is destined to be hit by lightning three times and still survive. All we need, Hand argues, is a firm grounding in a powerful set of laws: the laws of inevitability, of truly large numbers, of selection, of the probability lever, and of near enough.

continuar lendo   >>>>

Approximate Statistical Tests for Comparing Supervised Classification Learning Algorithms | Neural Computation | MIT Press Journals

June 25, 2017 no comments Posted in Análise de Dados

“This article reviews five approximate statistical tests for determining whether one learning algorithm outperforms another on a particular learning task. These test sare compared experimentally to determine their probability of incorrectly detecting a difference when no difference exists (type I error). Two widely used statistical tests are shown to have high probability of type I error in certain situations and should never be used: a test for the difference of two proportions and a paired-differences t test based on taking several random train-test splits. A third test, a paired-differences t test based on 10-fold cross-validation, exhibits somewhat elevated probability of type I error. A fourth test, McNemar’s test, is shown to have low type I error. The fifth test is a new test, 5 × 2 cv, based on five iterations of twofold cross-validation. Experiments show that this test also has acceptable type I error. The article also measures the power (ability to detect algorithm differences when they do exist) of these tests. The cross-validated t test is the most powerful. The 5×2 cv test is shown to be slightly more powerful than McNemar’s test. The choice of the best test is determined by the computational cost of running the learning algorithm. For algorithms that can be executed only once, Mc-Nemar’s test is the only test with acceptable type I error. For algorithms that can be executed 10 times, the 5 × 2 cv test is recommended, because it is slightly more powerful and because it directly measures variation due to the choice of training set.”

continuar lendo   >>>>

Do You Trust an Internet-Based service which takes more than 48 hours to answer your questions? – I don’t. Why? Because ..

June 23, 2017 no comments Posted in Publicando e Lendo

Do You Trust an Internet-Based service which takes more than 48 hours to answer your questions?

Do You Trust an Internet-Based service which takes more than 48 hours to answer your questions? - I don't. Why .... Because

some possible reasons

they are too big and making so much money that they don't care

they are growing too fast and don't have enough resources to cope with the demand

they are not making money proportional to the demand onto their support team

Anyway, you see, whatever are the possible reasons, they may explain but don't solve your problem, and how can you set some service dependent on them?

You don't, so keep distance.

Have You any other ideas on this?

Why Evolution is True

June 23, 2017 no comments Posted in Publicando e Lendo

Among the wonders that science has uncovered about the universe, no subject has sparked more fascination and fury than evolution. Yet in all the ongoing debates about creationism and its descendant, “intelligent design,” one element of the controversy is rarely mentioned: the evidence, the empirical truth of evolution by natural selection. And that evidence is vast, varied, and magnificent, drawn from a huge spectrum of scientific inquiry ranging from genetics, anatomy, and molecular biology to paleontology and geology.Why Evolution Is True is a succinct and accessible summary of the facts supporting Darwinian evolution. Scientists today are finding species splitting in two, observing natural selection changing animals and plants before our eyes, and discovering more and more fossils capturing change in the past-dinosaurs that have sprouted feathers, fish that have grown limbs. Jerry Coyne eloquently shows that evolution does not destroy the beauty of life but enhances it.

continuar lendo   >>>>

Como Instalar o R no Android  | Linux Journal

June 22, 2017 no comments Posted in R

Este mês, você também aprenderá a instalar o software de estatística R completo em um dispositivo Android móvel. Os dispositivos móveis são poderosos o suficiente para instalar esses pacotes, e com a ajuda de Marius Hofert e Kurt Hornik, você aprenderá a usar o Linux em um ambiente chroot no Android. Mesmo que você não esteja interessado em instalar o R, o processo de instalação do Linux no Android é fascinante e divertido! Marius e Kurt passam por todo o processo, incluindo informações sobre desbloqueio de bootloaders e dispositivos de rooting.

continuar lendo   >>>>