The issue of fair valuation of real estate has been a major challenge in many countries. However, in recent times a number of techniques and models have been introduced to promote fair valuation of housing units. It is in the light of the above, that this work developed, trained and applied three artificial neural network breadboards to evaluate over 3000 residential real estate transactions from two major cities in Nigeria. The strengths of this work are - it is comprehensive, practical oriented and leads to an indepth understanding of both real estate finance and a real-life application of artificial neural networks. It is a must-read for research students, academics and real estate practitioners; as it provides theoretical and empirical insights into real estate valuation, research methods and artificial intelligence systems.