How to Explain the Obtained Results?
The next important step after applying the PCA algorithm is to explain the results.
We can explore the correlation between each principal component and variable. This relationship is called factor loading. These statistics make it easier to understand the underlying dimension of the component.
We use the results obtained from the previous section.
C1, C2 and C3 are the main components that the PCA model inferred. The Y-axis contains the names of the columns from the dataset. As we can see, the first principal component correlates the most with SalePrice, OveralQual, and GarageArea. Let's assume that these are the main variables that determine the value of a home.
The second component is mainly responsible for the BedroomAbvGr and KitchenAbvGr indicators - the number of bedrooms or kitchens, the main spaces in the house.
The third component does not seem to be as obvious as the previous 2. It has the most pronounced variable LotArea and MSSubClass with a negative correlation.
Factor loading is one of the main methods that allows us to explore the components that created the PCA. Next, we will look at other methods.
Quiz
The result of the PCA on the dataset with the rating of various places (buildings) in the city is presented in the table below.
Tack för dina kommentarer!
Fråga AI
Fråga AI
Fråga vad du vill eller prova någon av de föreslagna frågorna för att starta vårt samtal
Awesome!
Completion rate improved to 5.26How to Explain the Obtained Results?
Svep för att visa menyn
The next important step after applying the PCA algorithm is to explain the results.
We can explore the correlation between each principal component and variable. This relationship is called factor loading. These statistics make it easier to understand the underlying dimension of the component.
We use the results obtained from the previous section.
C1, C2 and C3 are the main components that the PCA model inferred. The Y-axis contains the names of the columns from the dataset. As we can see, the first principal component correlates the most with SalePrice, OveralQual, and GarageArea. Let's assume that these are the main variables that determine the value of a home.
The second component is mainly responsible for the BedroomAbvGr and KitchenAbvGr indicators - the number of bedrooms or kitchens, the main spaces in the house.
The third component does not seem to be as obvious as the previous 2. It has the most pronounced variable LotArea and MSSubClass with a negative correlation.
Factor loading is one of the main methods that allows us to explore the components that created the PCA. Next, we will look at other methods.
Quiz
The result of the PCA on the dataset with the rating of various places (buildings) in the city is presented in the table below.
Tack för dina kommentarer!
