This feature allows users to define guidelines on how to interpret different attribute values
It allows the user to define how to categorize different values. For instance, when would a certain value be considered High, Medium or Low
For Data Source of type Raster, the legend item is based on pixel value. The value of the pixel is represented by x
For Data Source of type Vector, the legend item is based on the value of the expression which is usually a function of attribute values. For example, if we are computing population density, we could use the number of people attribute divided by the size of the area under consideration