Data preprocessing

[iiiMatt]

Data preprocessing

The titanic.csv should be preprocessed before the data analysis in the preprocessing.R file. Following steps should be taken:

• Salutation extraction: The salutation should be extracted from the name ("Mr.", "Mrs.", "Ms." etc...), later they will be used to infer the missing age of some people. Note that some of the salutation are the same contentwise (e.g. "Ms." and "Miss."). The salutation "Master" describes a young boy. The extracted salutation should be saved into a new variable called "Salutation".
• Variable encoding: The variables „Survived“, „Sex“ and „Embarked“ should be encoded into factors and "Pclass" into an ordered-factor (1 > 2 > 3).
• Age inferrence: Using the newly created "Salutation" variable the age should be inferred, for example by taking the mean or the median for each salutation.
• Cabin data extraction: The following data should be extracted from the variable "Cabin":
  • "Port" or "Starboard": if the "Cabin" is odd it's on the "Port" otherwise it's on the "Starboard"
  • "Deck": leading letter of the cabin number. In case of missing letter it should be NA
• Cleanup: Remove the variables "PassengerID", "Name", "Ticket" and "Cabin" from the dataset.
• Saving: The final dataset should be saved in a new file.

[miamacarena]

1. Should we merge the code regarding preprocessed data to the main branch if it is done? Or should we just upload another rscript?

2. Also I have seen some points that are stil missing if Maté's Code is the recent one: (I created a checklist so the ones who are responsible for the task can tick it off)

Finish up the code:

• [x] factorized survived, sex, embarked
• [x] factorize Pclass ORDERED
• [x] get rid of the variables passenger ID, name, ticket, Cabin