Over the recent decades, increasing the cost of drilling and completion projects in the hydrocarbon reservoirs has prioritized the demand for reducing risks of these projects for engineers. Therefore, the task of providing enough and precise information on the geomechanical environment and in turn developing stable modeling of the oilfield is performed by petroleum exploration engineers. Determination and evaluation of geomechanical parameters and anisotropy of wells in the oil and gas reservoirs are carried out using core, well logging, well testing, and pressure data. Among these methods, well logging is of great importance owing to the high-speed and low-cost data acquisition. Well logging data used for determination of geomechanical parameters and evaluation of anisotropy, dipole shear sonic (DSI) and imager data along with petrophysical data are more common. Due to the high cost of DSI in the oil and gas wells, only a few DSI conducted in the selected wells in an oilfield. Therefore, evaluation of anisotropy and determination of geomechanical parameters in the wells lacking DSI without using estimation methods is inevitable. This would be more important in the reservoir study and also for a safe drilling, which require a good understanding of anisotropy and geomechanical parameters. In this thesis, it has been attempted to estimate DSI graphs in the oil wells lacking this type of data (three wells in one of the southern Iranian oilfields) using intelligent methods based on Artificial Neural Network (ANN) to estimate geomechanical parameters for wells lacking DSI and in turn to evaluate the anisotropy around the wells. The estimation of DSI data using ANN has been carried out in two steps: in the first step, a new model based on ANN has been developed using common well logging data (as an inlet) and DSI data (as a target). Since a good petrological correlation between those three wells with this well was found, the DSI data were estimated using the developed model in other wells. In this study, a two-layer ANN was adopted to estimate DSI parameters. In addition, a single-layer ANN was utilized for estimation of shear wave velocity. The results on the evaluation of anisotropy of three wells using estimation data indicated the maximum horizontal stress as N55E, N25E, and N15E, respectively. To validate these results, they were compared with the imager results, which showed the maximum horizontal stress in the three wells as N32E, N20E, and N20E, respectively. Comparing these results with multivariable analysis and validation with ANN confirms that ANN has been well able to estimate DSI parameters. As well, the estimated shear wave velocity was used for evaluation of geomechanical modules. To validate the results, the velocity estimation was carried out in the wells with DSI data. The comparison of estimated shear wave velocity with real value confirmed the true model applied for estimation.