Fluorescence is usually measured in relative fluorescence units (RFU), because the amount of light detected depends on many factors. The slope of a plot of fluorescence intensity (RFU) versus protein concentration (in mg/mL, for example) would have "units" of RFU/(mg/mL). Unless you have some standards of known hydrophobicity to compare this measurement with, the units don't mean anything because the fluorescence measurement is relative.

I don't know whether there is any standard for surface hydrophobicity, but one can imagine creating one by using certain readily reproduced protein samples of varying hydrophobicities (based on x-ray crystal structures perhaps) and creating a relative scale with them. The least hydrophobic could be set to 1 and the most to 10, or something like that. You would have to use these standards every time you made the measurement with a new protein.

Another approach would be to use a ratiometric fluorescent dye, for which the ratio of the fluorescence intensities at two emission wavelengths was related to surface hydrophobicity. Using the ratio eliminates the relative nature of fluorescence intensity readings. Some standard proteins for comparison would still be needed at first to establish the scale, but once that was done, only the ratio would be needed.

Thank you very much for the answer!