A One-hot-encoded column turns categorical variables into a form that could be provided to ML algorithms to do a better job in prediction. It turns categorical values into 0s or 1s. The values are discrete.

A dense feature inspects the result of a feature column and inspects categorical columns by transforming it to an indicator column first. It also specifies the absence of data by putting a 0 if data is absent or cannot be represented numerically. This can be useful because missing data can mess up a model.

Logistic Regression Histogram:

Boosted Tree Histogram:

Both together:

The boosted tree histogram has the mortality rate as being higher.

The curve above shows that the model is slightly overfitting, but the area under the curve takes up the majority of the graph which is a positive.

The plots below show that age and sex were both extremely high factors in choosing who would die or not. Paying more for a ticket also seems as if your rate of survival was much higher.

For me, Sex and Fare were the most important.