We describe the variation of the accumulated coronal helicity derived from the magnetic helicity flux through the photosphere in active region (AR) NOAA 10365, where several large flares and coronal mass ejections (CMEs) occurred. We used SOHO/MDI full-disk line-of-sight magnetograms to measure the helicity flux, and the integral of GOES X-ray flux as a proxy of the coronal energy variations due to flares or CMEs. Using the linear force-free field model, we transformed the accumulated helicity flux into a time sequence of the force-free parameter α accounting for flares or CMEs via the proxy derived from GOES observations. This method can be used to derive the value of α at different times during the AR evolution, and is a partial alternative to the commonly used match of field lines with EUV loops. By combining the accumulated helicity obtained from the observations with the linear force-free theory, we describe the main phases of the emergence process of the AR, and relate them temporally with the occurrence of flares or CMEs. Additionally, a comparison with the loop-matching method of fixing alpha at each time independently shows that the proposed method may be helpful in avoiding unrealistic or undetermined values of alpha that may originate from an insufficient quality of the image used to identify coronal loops at a given time. For the relative intensity of the considered events, the linear force-free field theory implies that there is a direct correlation between the released energy on the one hand and the product of the coronal helicity with the variation of α due to the event on the other. Therefore, the higher the value of the accumulated coronal helicity, the smaller the force-free parameter variation required to produce the same decrease in the free energy during the CMEs.