Velocity anisotropy is an important parameter in seismic analyses, especially for deviated wells in offshore fields when the velocity anisotropy effect must be corrected to obtain a realistic sonic log response equivalent to seismic data for small incident angles. However, there is no method for an anisotropy correction when only standard log data from a single well are available. We compiled experimental data of core velocity anisotropy from several recent publications to explore heuristic relationships between anisotropy parameters and general well log information and found a clear trend of velocity anisotropy in the crossplot domain between velocity and porosity. Subsequently, we developed a method to estimate Thomsen's anisotropy parameters ε and γ by a regression in this domain. In this paper, we first show an application result of the method to demonstrate the significance of the correction. Then we summarize how we developed the method. We also discuss an alternative approach using neural networks. Finally, we compare our method to one proposed by Li (2002).