What is a feature set?
A feature quantity is a numerical value of features.
For example, if you want to predict the price of a house, the characteristics of the building itself (number of floors, floor area, age of the building, etc.), the prefecture or city other than the building, the information on the nearest station, the distance from the nearest station, and the age of the building are also important features.
And the data that quantifies the distance from the nearest station (3 minutes walk) and the age of the building (15 years) are features.
Let's take a look at the actual data.
In the following image, everything except price is a feature value. These features are called explanatory variables.
Also, since price is the variable we want to predict, this variable is called the objective variable.
If a human being predicts the rent of a rental house, etc., even a person outside the real estate industry may be able to understand sensibly that the price will fall if the house is far from the station, or if the building is old.
And, in general, when the person who predicts is the person of the real estate industry, it is thought that there are many cases where it has wider and deeper knowledge than the person who is not so.
If you have broad and deep knowledge, you can imagine that there are many cases where the price you forecast is also more accurate.
This kind of knowledge is called domain knowledge, and it is very important in creating machine learning models.