This video discusses how to detect outliers using box plots, scatter plots, and z-scores. It also emphasizes techniques for handling outliers as part of data scrubbing or cleaning, specifically through z-score methods. This content builds on the previous video about data scrubbing.
Thresholds are:
Z-score = 3: Data points with a z-score greater than 3 are typically considered outliers on the high end.
Z-score = -3: Data points with a z-score less than -3 are considered outliers on the low end.