Unconventional hydrocarbon-bearing reservoirs are known for their challenging complexity, that requires more advanced approaches and methods to solve. Advanced methods can help geophysicists significantly to develop a strong understanding of the characteristics and properties of such reservoirs. As an example, from Oman, Amin formation is a major tight gas fields in the north of the Sultanate. The reservoir is located at a depth ranging from 4800 to 5000 m. Conventional methods have failed to provide a reasonable mapping of the Amin formation especially with the small number of wells penetrating the formation. Seismic records only contrast in rock impedances and covers large areas with lower resolution. While well logging data contains a highly resolution recorded properties, but only in well location. The objective of this thesis is to integrate well data with seismic in order to develop an understanding of the reservoir quality of the tight Amin Formation, using machine intelligence, namely, Artificial Neural Network (ANN). The ANN was used for predicting missing logs (Delta-T share) for well-1. Using ANN, porosity, permeability, water saturation and gamma ray maps were predicted along Amin horizon. Predicted maps indicate low porosity and permeability of Amin formation. The geological information about the formation have supported the derived maps. Moreover, an increase in the porosity was observed around the faulted and fractured zones. In addition, gamma ray map shows dominating range of natural radiation between (0-65 GAPI) which is an indicator for the sand. Using self-organizing ANN supported with six wells, Amin horizon was divided into different segments. With the help of well data (well 6 and well 2), water saturated sand was identified. Hydrocarbon-saturated sand was identified based on information from well 1,5 and 4. As a recommendation, involving more wells in the Artificial Neural Network will provide much-detailed maps with higher level of confidence in the interpretation.