Institutional Repository
Technical University of Crete
Technical University of Crete
EN
|
EL
| URI: | http://purl.tuc.gr/dl/dias/CF51598E-E9B2-422C-A0E9-54BC1DC7CF2B | ||
| Year | 2021 | ||
| Type of Item | Diploma Work | ||
| License |
|
||
| Bibliographic Citation | Georgios Tsouvalakis, "US housing market forecast through ANFIS (Adaptive Neuro-Fuzzy System)", Diploma Work, School of Production Engineering and Management, Technical University of Crete, Chania, Greece, 2021 https://doi.org/10.26233/heallink.tuc.88399 | ||
| Appears in Collections |
Home sales in are considered to be one of the key factors influencing a country's economy. For this reason, forecasting models have been developed, where they serve us in making decisions on real estate investments and even in avoiding possible financial risks. There is a rapid increase in sales over the years, due to population growth which results in higher demand for housing. Also, the increase of productivity and the automation of many works, show a reduction of the time and the cost of construction, where it favors in the reduction of the delivery time and the immediate service of the customer. In the present project, the forecast of real estate sales is attempted, based on a neuro-fuzzy model. The model consists of an ANFIS (Adaptive Neuro-Fuzzy Inference System) controller to forecast, based on the sales data. In addition, the success rate of the model is determined by the exported MATLAB diagrams as well as the RMSE (Root Mean Square Error) error values that occurred after the end of the process. In conclusion, the model reflects the dynamics of daily real estate sales that can be obtained from an ANFIS system and help predict short-term sales trends.
Copyright © DIAS 2013
