To put it simply, machine learning is all about learning and predicting data.
What is learning data? Being able to predict the resulting data based on the causative data.
That is, ** It becomes possible to predict the result (objective variable) by the cause (explanatory variable). ** **
When you think about the test in the human world, it often happens that the same question as the question book appears in the test, so I think that you may want to memorize the question book in its entirety and hope for the test. However, this is a study method with no applied power, and the same can be said for machine learning.
** It is natural that the result (answer) can be predicted using the cause (answer). ** **
This is called overfitting.
Be aware that this state is ** no deal with humans or machine learning **.