ConverSight interview question

What is the best data structure to use to remove the repeated data while working with data?