One of the important purposes of interpreting seismic data is the petrophysical evaluation of zones of a reservoir, which is useful for extracting important information about the structure and lithology of the reservoir and improves seismic interpretation. Determination of reservoir electrical facies plays an important role in the petrophysical evaluation of reservoir zones. The information extracted from well drilling diagrams can be classified as electrical facies. In this research, among various supervised learning strategies, a random forest (RF) algorithm is used, which is a set of decision trees, and is generally taught through bootstrap collection. The advantage of the decision tree algorithm is that it gives us a small measure of the importance or redundancy of each feature in the learning process. To do this, a monitoring learning algorithm that utilizes prior information known to the interpreter is implemented using seismic properties to estimate seismic facies. Supervised learning The instructional data set in the supervised algorithm includes input data and labels that contain input response values. First, well logs are performed to map facies using random forest (RF) algorithm. Then, after establishing communication and coordination between well data and seismic data, the determined facies are generalized to the seismic section using seismic markers.