# Sensitivity Analysis of Machine Learning Models for the Mass Appraisal of Real Estate. Case Study of Residential Units in Nicosia, Cyprus

^{1}

^{2}

^{*}

## Abstract

**:**

## 1. Introduction

#### 1.1. Background of the Study

#### 1.2. State of the Art

## 2. Comparable Evidence and Methods

#### 2.1. Database, Pre-Processing, Methods and Performance Metrics

- Unit Enclosed extent, which is the Internal Area in m${}^{2}$ (IntArea).
- The Unit covered extent, which is the Area of covered verandahs in m${}^{2}$ (CovVer).
- The Unit uncovered extent, which is the Area of uncovered verandahs in m${}^{2}$ (UnCovVer).
- Parcel extent, that is the Area of parcel (or plot) in m${}^{2}$ (ParcExt).
- The Built Years, calculated as the difference among the date the transaction happened and the date the building was constructed, in years (BuiltYrs).
- The Unit condition code (Cond), that denotes the condition of the building, and takes values from 1 (best condition) to 4 (worst condition).
- The Unit’s view code (View), which denotes the view of the unit, with values from 1 (best view) to 4 (worst view).
- The Unit’s class code (Class), denoting the class of the building. It takes Values from 1 (best class) to 4 (worst class).
- Density (Dens), as the maximum allowed density (built m${}^{2}$, over plots m${}^{2}$) of the specific district.

#### 2.2. Error Metrics

#### 2.3. Anomaly Detection

Algorithm 1:Anomaly Detection |

#### 2.4. Machine Learning Methods

Algorithm 2:Step-wise, Higher Order Regression |

## 3. Results

#### 3.1. Regression Analysis

#### 3.2. Sensitivity Analysis

#### 3.3. How Much Data Is Big Enough?

#### 3.4. Prediction Formula

## 4. Discussion

#### Remote Sensing Integration in Mass Appraisals

## 5. Conclusions

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## Appendix A

#### Appendix A.1. Prediction Formula with 100 terms (MAE = 19694€)

## References

- Makridakis, S.; Bakas, N. Forecasting and uncertainty: A survey. Risk Decis. Anal.
**2016**, 6, 37–64. [Google Scholar] [CrossRef] [Green Version] - Bakas, N.P. Numerical Solution for the Extrapolation Problem of Analytic Functions. Res.
**2019**, 2019, 1–10. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Jiang, F.; Jiang, Y.; Zhi, H.; Dong, Y.; Li, H.; Ma, S.; Wang, Y.; Dong, Q.; Shen, H.; Wang, Y. Artificial intelligence in healthcare: past, present and future. Stroke Vasc. Neurol.
**2017**, 2, 230–243. [Google Scholar] [CrossRef] [PubMed] - Pozna, C.; Antonya, C. Issues about autonomous cars. In Proceedings of the 2016 IEEE 11th International Symposium on Applied Computational Intelligence and Informatics (SACI), Timisoara, Romania, 12–14 May 2016; pp. 13–18. [Google Scholar]
- Gupta, S.; Sharma, A.; Abubakar, A. Artificial intelligence–driven asset optimizer. In Proceedings of the SPE Annual Technical Conference and Exhibition, Dallas, TX, USA, 24–26 September 2018. [Google Scholar]
- De Swarte, T.; Boufous, O.; Escalle, P. Artificial intelligence, ethics and human values: The cases of military drones and companion robots. Artif. Life Robot.
**2019**, 24, 291–296. [Google Scholar] [CrossRef] - Olson, C.; Levy, J. Transforming marketing with artificial intelligence. Appl. Mark. Anal.
**2018**, 3, 291–297. [Google Scholar] - Zadeh, L.A.; Tadayon, S.; Tadayon, B. System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform. U.S. Patent App. 15/919,170, 3 December 2018. [Google Scholar]
- Cave, S.; Dihal, K. Ancient dreams of intelligent machines: 3,000 years of robots. Nature
**2018**, 559, 473–475. [Google Scholar] [CrossRef] - MAYOR, A. Gods and Robotss: Myths, Machines, and Ancient Dreams of Technology; Princeton University Press: Princeton, NJ, USA, 2018. [Google Scholar] [CrossRef]
- Dimopoulos, T.; Bakas, N. An artificial intelligence algorithm analyzing 30 years of research in mass appraisals. Reland Int. J. Real Estate & Land Plan.
**2019**, 2, 10–27. [Google Scholar] - Dimopoulos, T.; Tyralis, H.; Bakas, N.P.; Hadjimitsis, D. Accuracy measurement of Random Forests and Linear Regression for mass appraisal models that estimate the prices of residential apartments in Nicosia, Cyprus. Adv. Geosci.
**2018**, 45, 377–382. [Google Scholar] [CrossRef] [Green Version] - Merriam Webster; Merriam Webster: Miami, FL, USA, 2016.
- Worzala, E.; Lenk, M.; Silva, A. An exploration of neural networks and its application to real estate valuation. J. Real Estate Res.
**1995**, 10, 185–201. [Google Scholar] - Brynjolfsson, E.; Mitchell, T.; Rock, D. What Can Machines Learn, and What Does It Mean for Occupations and the Economy? Aea Pap. Proc.
**2018**, 108, 43–47. [Google Scholar] [CrossRef] [Green Version] - Pagano, D. Machine Learning Will Replace Tasks, Not Jobs, Say MIT Researchers. MIT News, 26 June 2018. [Google Scholar]
- Bryson, J.J. Robots should be slaves. In Close Engagements with Artificial Companions: Key Social, Psychological, Ethical and Design Issues; University of Bath: Bath, UK, 2010; pp. 63–74. [Google Scholar]
- Agrawal, S.; Agrawal, J. Survey on anomaly detection using data mining techniques. Procedia Comput. Sci.
**2015**, 60, 708–713. [Google Scholar] [CrossRef] [Green Version] - Dimopoulos, T.; Moulas, A. A proposal of a mass appraisal system in Greece with CAMA system: Evaluating GWR and MRA techniques in Thessaloniki Municipality. Open Geosci.
**2016**, 8, 675–693. [Google Scholar] [CrossRef] - Chan, A.P.; Abidoye, R.B. Advanced property valuation techniques and valuation accuracy: Deciphering the artificial neural network technique. Reland Int. J. Real Estate & Land Plan.
**2019**, 2, 1–9. [Google Scholar] - Arribas, I.; García, F.; Guijarro, F.; Oliver, J.; Tamošiūnienė, R. Mass appraisal of residential real estate using multilevel modelling. Int. J. Strateg. Prop. Manag.
**2016**, 20, 77–87. [Google Scholar] [CrossRef] [Green Version] - Ciuna, M.; Milazzo, L.; Salvo, F. A mass appraisal model based on market segment parameters. Building
**2017**, 7, 34. [Google Scholar] [CrossRef] - Chica-Olmo, J.; Cano-Guervos, R.; Chica-Rivas, M. Estimation of housing price variations using spatio-temporal data. Sustainability
**2019**, 11, 1551. [Google Scholar] [CrossRef] [Green Version] - Cyprus New General Valuation 2013. 2013. Available online: https://portal.dls.moi.gov.cy/en-us/Pages/New-General-Valuation-as-at-1-1-2013.aspx (accessed on 1 June 2019).
- Lelo, K.; Tomassi, S.M.F. Urban inequalities in Italy: A comparison between Rome, Milan and Naples. Entrep. Sustain. Issues
**2018**, 6, 939–957. [Google Scholar] [CrossRef] - NCSS Statistical Software. Appraisal Ratio Studies; NCSS: Kaysville, UT, USA, 2015. [Google Scholar]
- Breiman, L. Random forests. Mach. Learn.
**2001**, 45, 5–32. [Google Scholar] [CrossRef] [Green Version] - Sadeghi, B. DecisionTree.jl. 2013. Available online: https://github.com/bensadeghi/DecisionTree.jl (accessed on 1 June 2019).
- Xu, B.; Chen, T. XGBoost.jl. 2014. Available online: https://arxiv.org/abs/1603.02754 (accessed on 1 June 2019).
- Bezanson, J.; Edelman, A.; Karpinski, S.; Shah, V.B. Julia: A fresh approach to numerical computing. Siam Rev.
**2017**, 59, 65–98. [Google Scholar] [CrossRef] [Green Version] - Gevrey, M.; Dimopoulos, I.; Lek, S. Review and comparison of methods to study the contribution of variables in artificial neural network models. Ecol. Model.
**2003**. [Google Scholar] [CrossRef] - Olden, J.D.; Jackson, D.A. Illuminating the “black box”: A randomization approach for understanding variable contributions in artificial neural networks. Ecol. Model.
**2002**. [Google Scholar] [CrossRef] - Sagi, O.; Rokach, L. Ensemble learning: A survey. Wiley Interdiscip. Rev. Data Min. Knowl. Discov.
**2018**, 8, e1249. [Google Scholar] [CrossRef] - Oreskes, N.; Shrader-Frechette, K.; Belitz, K. Verification, validation, and confirmation of numerical models in the earth sciences. Science
**1994**. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Lybeck, E. The black swan: The impact of the highly improbable. arXiv
**2017**, arXiv:1011.1669v3. [Google Scholar] - Bird, R.; Slack, E.; Guevara, M. Real Property Taxation in the Philippines. In International Handbook of Land and Property Taxation; Edward Elgar Publishing: Cheltenham, UK, 2013. [Google Scholar] [CrossRef]
- Autor, D.H. Why Are There Still So Many Jobs? The History and Future of Workplace Automation. J. Econ. Perspect.
**2015**. [Google Scholar] [CrossRef] [Green Version] - Ismail, S.; Buyong, T. Residential Property Valuation Using Geographic Information System; Bulgarian Geoinformation Company: Ovcha Kupel, Bulgaria, 1998. [Google Scholar]
- Frey, C.B.; Osborne, M.A. The future of employment: How susceptible are jobs to computerisation? Technol. Forecast. Soc. Chang.
**2017**. [Google Scholar] [CrossRef] - Dimopoulos, T.; Labropoulos, T.; Hadjimitsis, D.G. Comparative analysis of property taxation policies within Greece and Cyprus evaluating the use of GIS, CAMA, and remote sensing techniques. Proc. SPIE Int. Soc. Opt. Eng.
**2014**, 9229, 92290O. [Google Scholar] - Nayak, S.; Zlatanova, S. Remote Sensing and GIS Technologies for Monitoring and Prediction of Disasters; Springer: Berlin, Germnay, 2008. [Google Scholar]
- LIU, X.S.; Zhe, D.; WANG, T.L. Real estate appraisal system based on GIS and BP neural network. Trans. Nonferrous Met. Soc. China
**2011**, 21, s626–s630. [Google Scholar] [CrossRef] - Lindgren, D. Land Use Planning and Remote Sensing; Taylor & Francis: New York, NY, USA, 1984; Volume 2. [Google Scholar]

Sample Availability: The dataset was provided from the Department of Lands and Surveys. |

Methods | $\mathit{\rho}$ | MAE | RMSE | MAPE | MAXAPE | SR | $\mathit{\alpha}$ | COD |
---|---|---|---|---|---|---|---|---|

Train Set | ||||||||

Random Forests | 0.914 | 17931.100 | 28854.237 | 0.111 | 1.307 | 1.031 | 0.739 | 10.778 |

Gradient Boosting | 0.992 | 2630.784 | 8923.668 | 0.016 | 0.441 | 1.002 | 0.983 | 1.753 |

Linear Regression | 0.863 | 24546.300 | 34745.422 | 0.151 | 0.550 | 1.027 | 0.746 | 14.703 |

Non-Linear Regression | 0.880 | 23520.570 | 32700.793 | 0.146 | 1.100 | 1.032 | 0.775 | 14.197 |

Test Set | ||||||||

Random Forests | 0.877 | 20817.165 | 27950.722 | 0.134 | 0.802 | 1.040 | 0.753 | 12.950 |

Gradient Boosting | 0.803 | 24485.519 | 35946.437 | 0.151 | 1.092 | 1.009 | 0.776 | 15.017 |

Linear Regression | 0.858 | 22977.825 | 30047.707 | 0.146 | 0.506 | 1.025 | 0.789 | 14.279 |

Non-Linear Regression | 0.862 | 22525.779 | 29500.974 | 0.144 | 0.552 | 1.032 | 0.761 | 13.984 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Dimopoulos, T.; Bakas, N.
Sensitivity Analysis of Machine Learning Models for the Mass Appraisal of Real Estate. Case Study of Residential Units in Nicosia, Cyprus. *Remote Sens.* **2019**, *11*, 3047.
https://doi.org/10.3390/rs11243047

**AMA Style**

Dimopoulos T, Bakas N.
Sensitivity Analysis of Machine Learning Models for the Mass Appraisal of Real Estate. Case Study of Residential Units in Nicosia, Cyprus. *Remote Sensing*. 2019; 11(24):3047.
https://doi.org/10.3390/rs11243047

**Chicago/Turabian Style**

Dimopoulos, Thomas, and Nikolaos Bakas.
2019. "Sensitivity Analysis of Machine Learning Models for the Mass Appraisal of Real Estate. Case Study of Residential Units in Nicosia, Cyprus" *Remote Sensing* 11, no. 24: 3047.
https://doi.org/10.3390/rs11243047