0002- supervised learning machine learning notes

Supervision algorithm means that we give the algorithm a data set that contains the correct answer, the purpose of the algorithm is to give more correct answers.

Regression

Prediction value output continuous

Classification

Discrete values ​​predicted output

Examples

• Question 1: Do you have a lot of inventory with a cargo of, you want to predict in the next three months, how many can you sell
  a problem due to the number of pieces to be predictive of goods sold is a continuous value, so this problem belongs return
• Question 2: Do you have a lot of users, you need to detect each client's account by a software to determine whether this account is invaded or destroyed
  account whether the invasion or destruction of discrete values is limited, so the question of the class issues