This paper reports on an ongoing effort to provide computational linguistic support to scholars making use of the writings from the Third Dynasty of Ur, especially those trying to link reports of financial transactions together for the purpose of social networking. The computational experiments presented are especially addressed to the problem of identifying proper names for the ultimate purpose of reconstructing a social network of UR III society. We describe the application of established supervised learning algorithms, compare its results to previous work using unsupervised methods and propose future work based upon these comparative results.