Abstract: The Ordos Basin is an important unconventional oil and gas production area in China. The Yanchang Formation Chang 7 reservoir is rich in shale oil and has great potential for exploration and development. Reservoir geomechanics evaluation is critical for guiding and realizing the efficient development of shale oil in Ordos Basin. Identifying the geomechanical characteristics of Chang 7 shale oil reservoir and establishing a quantitative prediction model are important bases for supporting reservoir fracturability evaluation and sweet spot interval optimization. In this study, typical Chang 7 reservoir core samples in southern Ordos Basin were collected, and geomechanical parameters such as elastic modulus, Poisson's ratio, present-day in-situ stresses were experimentally determined. Further, a geomechanical parameter prediction model based on logging data was constructed with the BP neural network to achieve quantitative evaluation of geomechanical parameters and fracturability in the Yanchang Formation Chang 7 shale oil reservoir. The results show that: 1) the reservoir geomechanical parameter prediction model based on BP neural network has high accuracy, and the error between prediction results and measured values ​​is small; 2) the key geomechanical parameters of Chang 7 shale oil reservoir are heterogeneous, elastic modulus is between 16.26 and 59.12 GPa, fracture toughness is between 0.2~1.2 MPa·m^0.5, the horizontal maximum principal stress and the horizontal minimum principal stress are between 20 and 43 MPa, 12 and 38 MPa, respectively; 3) the Chang 7 shale oil fracturability evaluation index F based on reservoir geomechanical parameters is constructed, and the reservoir grades are divided. According to the classification, the reservoirs are divided into four levels: Class I reservoirs F>2.00, Class II reservoirs 2.00>F>1.50, Class III reservoirs 1.50>F>0.10, and Class IV reservoirs F<1.00. The research results can provide scientific guidance for reservoir fracturing optimization design.