There are several issues you may encounter, in principle, and I hope you will be getting more qualified answers from expert theorists.

(i) DFT is a single particle theory. In most implementations or without appropriate correction schemes (in the hands of skilled people, I reckon...) the band gap comes out too small. That is a typical deficiency of DFT and well documented in the pertinent literature.

(ii) DFT calculates a total energy (per unit cell). The relationship with single particle band structure is pretty well understood today, I think, but nonetheless not entirely trivial. (After all, you're solving a variational problem, and the QP energies are associated Lagrange multipliers).

(iii) spin orbit interaction is an ingredient to the calculation which may turn out to be  important to get things quantitatively correct.

(iii) Optical absorption is a two particle (= electron-hole) process/excitation and hence in principle nontrivially connected with the single particle band structure.

(iv) Excitons come on top of that. The presence of excitonic interactions will shift the absorption to smaller energies than predicted by the single particle band structure.

(v) The peak in absorption is not necessarily given at the band edge. In the single particle approximation, you are roughly looking for |DME|2*JDOS(\hbar\omega), where "DME" is "dipole matrix element" and "JDOS(\hbar\omega)" is "joint density of states with energy difference \hbar\omega". In your calculation, you do have significant absorption around (and below) 2 eV, as you would expect from the band structure spaghetti.

(vi) The total oscillator strength "available" is independent on the interactions (the so-called f-sum-rule). The exciton(s) may acquire (a) considerable oscillator strength but then does so at the expense of other excitations.

(vii) In the bulk crystals, the exciton binding energies are WAY smaller (~ 1 order of magnitude) than what you mention. In a monolayer, the exciton is actually strongly confined in the direction perpendicular to the layer. This will indeed lead to a stronger exciton binding energy. I do not actually know whether that really leads to an increase by an order of magnitude.

(viii) My recommendations: do actively get in contact with experts not only in the DFT package but also their application to the specific material, and on calculating optical properties in particular.  [Also, if not already done, try reproducing calculations from other people's publications to see whether you get consistent results. All this to make sure you have the right starting point for your further investigations.]

P.S. Concerning bulk WSe2 (a structurally related compound) we have taken a look at single particle excitations and their "optical counterparts" in the (relatively old [-fashioned]) paper referred to below (no full text on RG, please visit PRB in case. For comparison, cross-check with the paper just preceding ours. Since then, at least the photoemission part was reported with better instrumental resolution in more recent work). It could serve as a useful reference, nevertheless, be aware that both single particle band structure and optical excitations are severely altered by the confinement to 2D in the single layer compound.

Article Occupied and unoccupied electronic band structure of WSe_ {2}

solution of the question in this paper

for error band gap must be used functional hybrid eg HSE code CASTEP or functional MBJ  code WIEN2k

Why did you discard the first peak around 2 eV of the optical absorption? With this I think the band structure and optical spectrum (i.e. imaginargy part of the dilectric function) are completely agree with each other.