✳flâneur — a map of the web's best reading

Decision_Trees.ipynb - Colaboratory

colab.research.google.com · 4 words · saved by 1 readers

This time we will use a clasic introductory dataset that contains demographic and traveling information for the Titanic passengers. The goal is to predict the survival of these passengers. We will only keep a few variables of interest and transform all of them to numeric variables. We will also drop some outliers. It looks like there might be some missing values of some of these variables, becasue they have different counts. Let's confirm Wow. There are a lot of missing values for the age variable. We would want to check and see if the missing cases are different in any way. For starters, are they more or less likely to die? Lets check. This is concerning. In a more thorough analysis of this data, we would study the missing-ness more deeply to see if the missing values are correlated with other attributes, which would potentially add bias to the results Also, we would consider options for filling in - imputing - the data. However, in order to discuss the things we want to discuss, we w

Google Colab Sign in

Explore this link on the map →

related reading