Abstarct: The sequence stratigraphy of science is the study of the generational relationship between sedimentary units that are used in exploration of hydrocarbon reserves to determine the source rock, cap rock and reservoir rock. By combining well data and seismic data, a more precise interpretation of the sequence of oil fields can be obtained. Interpretation of sequence strain on seismic sections due to the low quality of some of these sections or the parallel seismic horizons in them, it is time consuming and requires specialized people in this field. Therefore, this problem can be solved by using multi-level analysis and applying classification methods. For this purpose, firstly seismic data and well data were adapted in the time zone by check shot and the production of synthetic traces. In the second stage, the appropriate seismic attributes of sequence stratigraphy were identified by SFS method. These seismic attributs include the RMS domain, domain extensions, and domain strength indicators. In the third step, the depth of the boundary of the sequences was determined by the well information and used to train the K-closest neighboring algorithm and Adaboost algorithm. After ensuring the accuracy of the training algorithms, the entire 2D cross section of the wells was classified. According to the matrix matrix, the classification of sequences in a two-dimensional cross section with a quantitative error value is performed. In the final output, the classification of three third-order sequences (baxsed on well data) was detected. Finally, by multi-attribute analysis, effective porosity was classified by Adaboost algorithm and it was concluded that the first sequence had good effective porosity.