The Titanic dataset is one of the most famous ones out there. It’s based on the real-life tragedy that occurred decades ago when the Titanic ocean liner hit an iceberg and sunk, killing many people.
The dataset is a supervised learning one (classification) – supposedly for beginners but not as easy as advertised – where we are supposed to predict who survived. Of course, this is quite awkward given that this was a real life event.
My approach to this dataset was a bit different. I spent a lot of time analysing the dataset to see any patterns amongst the survivors. I also tried multiple different models to see if any stood out in particular. It seems that all of them gave similar results.
Photo credit : Wikipedia
