Abstarct: One of the goals of the seismic perceptions is presentation of model of the reservoir changes trend and anisotropy in the subsurface laxyers. However, well log data can provide more accurate estimates of reservoir changes, but the data is one dimensional and they are picked up for particular depth. Thus, if we would be able to relate the values of reservoir parameters that obtained from well data to seismic data, suitable two-dimensional or three-dimensional models will be provided for a wide range of studied field, which opens the door to many other studies. Interpretation of seismic facies analysis parameters, such as geometry, continuity, amplitude, frequency and velocity of subsurface laxyers which are extractable from reflective seismic data. Since the reflection of subsurface laxyers is dependent on rocks properties, seismic facies analysis can provide a proper analysis of subsurface laxyers changes trend. In this context, the aim of the present study is seismic facies analysis one of the oil fields by using neural network and support vector machine. For this purpose, we tried to use well data and seismic one of the fields of the black sea and made model of facies changes trend. This is the process using log data on porosity of 15 different wells and different seismic attribute that done of seismic traces of next to the well. Related seismic attribute with porosity logs selected by using linear regression and were removed high solidarity attributes with together. Finally by using simulation of Gaussian smoothing function, three-dimensional model of trend parameter variations in porosity was offered that had good adaptation with observed changes in reservoir. Also with the use of porosity log data, permeability and gamma radiation, we tried to predict observable sediment zone amounts in the well data. In this context, the nonlinear support vector machine with Gaussian kernel was used and was shown that, this machine according to the own unique capabilities can separate different sediment laxyers from each other