The function takes a numerical input and multiplies it by a scaling factor that is dependent on the input's sign. Positive inputs are scaled by a positive factor, while negative inputs are scaled by a negative factor.

