Well log data analysis plays an important role in petroleum exploration. It is used to identify the potential for oil production at a given source and so forms the basis for the estimation of financial returns and economic benefits. In recent years, many computational intelligence techniques such as backpropagation neural networks (BPNN) and fuzzy systems have been applied to perform the task. Support vector machines (SVMs) are new techniques and very few reports have been published in this application area. This paper presents the investigation and comparison of BPNN model with a SVM model on a set of practical well log data. Future directions of exploring of the use of SVM for improved results will also be discussed.