Exploring data #

Ultimately, any machine learning is just optimizing error in a graph, then deriving a function that fits to the data.

Read more about John Tukey

Iris sample data set #

• http://www.ics.uci.edu/~mlearn/MLRepository.html

Statistics definitions #

• ordinal means that the data can be ordered
• continuous means that the values are floating point

• \( \text{AAD} (x) \) is the absolute distance
• \( \text{MAD} (x) \) is the median dependent value

Visualization #