Transactional price is the result of some kind of free market game, and of independent decisions taken by the parties to the transaction. Prices depend on a number of factors, specific to the local real estate market. The impact of some factors is fixed, while others are dependent on the loca- tion of the property. Therefore, research into the determination of rules that would describe the relationship between the market price of the real estate, and its market characteristics, remain valid. The article presents the possibilities of applying linear artificial neural networks to real estate valuation. Using a database, the artificial linear neural network is developing a regression model, which produces the results that oscillate close to the market value of the property. The necessary condition is the creation of a database that is representative of the given real estate market.