Artificial intelligence

Abstract: The article deals with the problem of improving quality of real estate datasets which are created from the open sources of information. Main problems of such datasets are analyzed and some ways to solve them are suggested in this article. These solutions were tested on different algorithms for real estate valuation and the best ones were selected.

Keywords: real estate valuation, training dataset, artificial neural networks.

References:

1. Helbich M. et al. Data-driven regionalization of housing markets //Annals of the Association ofAmerican Geographers. – 2013. – Т. 103. – №. 4. – S. 871-889.
2. Zurada J., Levitan A., Guan J. A comparison of regression and artificial intelligence methods in a massappraisal context //Journal of Real Estate Research. -- 2011.
3. Baranzini A., Schaerer C. A sight for sore eyes: Assessing the value of view and land use in the housingmarket //Journal of Housing Economics. -- 2011. -- Т. 20. -- №. 3. - S. 191-199.
4. Kuşan H., Aytekin O., Özdemir İ. The use of fuzzy logic in predicting house selling price //Expertsystems with Applications. -- 2010. -- Т. 37. -- №. 3. -- S. 1808-1813.
5. Bourassa S., Cantoni E., Hoesli M. Predicting house prices with spatial dependence: A comparison ofalternative methods //Journal of Real Estate Research. – 2010.
6. Kauko T. A comparative perspective on urban spatial housing market structure: Some more evidence oflocal sub-markets based on a neural network classification of Amsterdam //Urban Studies. – 2004. –Т. 41. – №. 13. – S. 2555-2579.