This is a pitch presentation about the Titanic Survival Test application.

I made this application based on this kaggle exercise: https://www.kaggle.com/c/titanic

I performed an analysis on the data obtained at Kaggle so we can know which fields are most related to the Survived field.

Some fields on the following plot are missing because they added too much complexity and didn't affected the Survived field after all.

The application is hosted at shiny apps (https://jonathanglima.shinyapps.io/CourseraDevelopingDataProductsProject) and it uses shiny to run predictions on a model that I created before.

It loads up the model (that's why it takes a few seconds at the beggining) and afterwards it runs a prediction on the model using the form filled data.

As we can see in the pairs graphic unfortunately only few people survived the Titanic accident. We can also see that most of the decision to survive or not were based on Sex, Class and slightly the Age.

That is also clear when you use the application, which clearly have drastical changes when you change the Sex and the Class.

Checking out the accident "per se" we can understand that due to lack of emergency boats the strategy that was adopted was that women and children went first. The class affected the strategy because third class was at the lower part of the ship. Since it was the first part to flood, that's the reason why there were so many deaths on the third class.