From 23736578651633d3e7e6e94c3df87271af9c7bcd Mon Sep 17 00:00:00 2001 From: vicarizmendi Date: Mon, 6 Jun 2016 18:34:13 +0200 Subject: [PATCH] Update index.Rmd Formula of majority vote accuracy corrected. * $10\times(0.7)^3(0.3)^2 + 5\times(0.7)^4(0.3) + (0.7)^5$ instead of * $10\times(0.7)^3(0.3)^2 + 5\times(0.7)^4(0.3)"""^2""" + (0.7)^5$. Checked the new formula reflects the result given of 83.7% majority vote accuracy. --- 08_PracticalMachineLearning/025combiningPredictors/index.Rmd | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/08_PracticalMachineLearning/025combiningPredictors/index.Rmd b/08_PracticalMachineLearning/025combiningPredictors/index.Rmd index 958b2c4d8..b9507fd77 100644 --- a/08_PracticalMachineLearning/025combiningPredictors/index.Rmd +++ b/08_PracticalMachineLearning/025combiningPredictors/index.Rmd @@ -67,7 +67,7 @@ BellKor = Combination of 107 predictors Suppose we have 5 completely independent classifiers If accuracy is 70% for each: - * $10\times(0.7)^3(0.3)^2 + 5\times(0.7)^4(0.3)^2 + (0.7)^5$ + * $10\times(0.7)^3(0.3)^2 + 5\times(0.7)^4(0.3) + (0.7)^5$ * 83.7% majority vote accuracy With 101 independent classifiers