In the last two to three decades, the use of seismic attributes for reservoir characterization and modeling has grown exponentially. Now, a dozen or more attributes are often extracted from seismic data to predict reservoir properties. Meanwhile, an increasing trend of acquiring more wireline logs provides more and more data to describe reservoir properties. Both statistical methods and artificial neural networks (ANNs) are often used to extract information and make predictions. Although statistical methods and ANNs provide powerful tools for geoscience data integration, they also have pitfalls. We present principal component analysis (PCA) and ANNs for facies classifications and porosity prediction. We also show the use and limitations of these methods and the importance of integrating the geologic and petrophysical knowledge.