Forecasting Rental Values of Residential Properties: A Neural Network Model Approach

Olalekan Oshodi, Ifije Ohiomah, Tawakalitu Odubiyi, Clinton Aigbavboa, Wellington Thwala

Research output: Chapter in Book/Report/Conference proceedingChapterpeer-review

1 Citation (Scopus)

Abstract

The current study intends to use the neural network (NN) algorithm for modelling and forecasting of rental values of residential properties located in Cape Town, South Africa. Data relating to property attributes and its rental value were collected. Neural network algorithm was applied in this study. The collected data was divided into two parts. The first part was used for the development of the model. Subsequently, the developed model was used to generate the forecast of rental values of residential properties. For the second part of the data, the accuracy of the model was evaluated by comparing the predicted class and actual class. Experimental results gave an accuracy of 66.67% for the test dataset. It was also found that floor area has the most significant impact on the rental value of residential properties within the study area. This study demonstrates that the neural network algorithm could be applied to real-world investigations focused on the prediction of rental values of residential properties.

Original languageEnglish
Title of host publicationAdvances in Science, Technology and Innovation
PublisherSpringer Nature
Pages309-313
Number of pages5
DOIs
Publication statusPublished - 2021

Publication series

NameAdvances in Science, Technology and Innovation
ISSN (Print)2522-8714
ISSN (Electronic)2522-8722

Keywords

  • Classification
  • Forecasting
  • Modelling
  • Property economics
  • Rental value

ASJC Scopus subject areas

  • Architecture
  • Renewable Energy, Sustainability and the Environment
  • Environmental Chemistry

Fingerprint

Dive into the research topics of 'Forecasting Rental Values of Residential Properties: A Neural Network Model Approach'. Together they form a unique fingerprint.

Cite this