The colon separate a dopant, whose concentration is too low to be given the same way as the crystal components, and may also vary more. It's vital as this defines the type (P or N) of the semiconductor, and must also be the dopant that defines the light-emitting transition.
I didn't check your semiconductor, but in GaP LED, a dopant introduces a deep level in the band gap. It's filled with electrons injected by current from the conduction band, and when it empties itself to the valence band, it emits a photon. The dopant defines the deep level, the height of the optical transition, hence the LED colour.
The subscripts indicate a chemical compositions, just like H2O tell 2 atoms H for 1 atom O. Here, they're adjusted, hence the variables a and b.
This is possible because Y and Gd have similar properties, especially the valence. Same for Al and Ga.
The trick with these crystals is that during their production by epitaxy, you can adjust the composition to obtain the property you want, which is often the height of the band gap (but it changes the refractive index as well for instance).
If you change only one atom pair, the lattice parameter may vary, preventing a good epitaxy (GaaAl1-aAs is an exception). But by adjusting the proportions of two atom pairs at the same time, you can obtain the bandgap you want AND the lattice parameter.
This was first used at semiconductor lasers, where a double heterojunction concentrates both the charge carriers and the light. Apparently LED are to use them as well.
Wiki must have an article about laser diodes, then heterojunctions will tell more.