Ali Sajedian - MBA, Fanpardazan Institute of Higher Education, Iran
Mojdeh Rahmani - Master of Marketing Manager, Islamic Azad University, Iran
Nima Rahmani - Bachelor of IT, Islamic Azad University, Iran

One of the most important economic concerns in any family is house purchasing, in which real state agencies as the intermediary between buyer and seller plays the most important role in pricing. Due to the dependence of housing prices on various parameters and the lack of experts in all of these firms, this estimate has been tough and sometimes put in trouble. Therefore, finding a solution to accurately estimate housing prices can be one of the main problems. In this article, by applicable the deep learning method and using a database in the state of California which was prepared by experts (published on Github website), the housing price was estimated. The available information included textual information as well as photographs taken of the house, which made modeling and implementing all of these parameters difficult. The collected features extracted from textual and visual features were fed to a convolutional neural network model that estimates the house price as its single output. Our work differs markedly from former research that has made use of images to price houses. The presented model and the obtained results show the high accuracy of this method that can be used similarly to all housing companies in all cities of the world. Also, real state agencies can easily use these models to estimate property prices

Real Estate, Deep Learning, Convolutional Neural Network (CNN), California