Coursera Machine Learning Study notes (ii)

-Supervised learning

For supervised learning let's look at an example, which is an example of a house price forecast. The horizontal axis of the figure shows the floor space, and the ordinate indicates the price of the house transaction. Each fork in the figure represents a house instance.

Now, we want to be able to predict the price of a house with a housing area of 750 square feet. The simple method is to draw an appropriate line based on the distribution of these data points, and then predict them according to the line. Of course, in this House price forecast example, a two-time function is more suitable for the distribution of existing data. Therefore, we may prefer to use this two-time function curve to make house price predictions.

Therefore, we call this study called supervised learning, because we gave the algorithm the correct results beforehand, that is, all the variables we observed.

The above-mentioned problem is also called regression problem, we predict the value is a continuous value.

Next, let's look at an example of supervised learning:

Suppose we want to predict whether a patient's tumour is malignant, depending on the size of the patient's tumour:

Of course, sometimes we use more than one variable, such as the age of the patient, the size and shape of the tumour, and so on.

In the picture, the circle represents benign and the fork is malignant, and the problem we want to learn becomes the division of benign tumors and malignant tumors.

This problem is also called classification problem, the classification of the use of discrete values. We want to use this algorithm to classify the data into different kinds.

Coursera Machine Learning Study notes (ii)