Combining BP with PSO algorithms in weights optimisation and ANNs training for mass appraisal of properties

Abstract

Purpose The paper aims to investigate the application of particle swarm optimisation and back propagation in weights optimisation and training of artificial neural networks within the mass appraisal industry and to compare the performance with standalone back propagation, genetic algorithm with back propagation and regression models. Design/methodology/approach The study utilised linear regression modelling before the semi-log and log-log models with a sample of 3,242 single-family dwellings. This was followed by the hybrid systems in the selection of optimal attribute weights and training of the artificial neural networks. Also, the standalone back propagation algorithm was used for the network training, and finally, the performance of each model was evaluated using accuracy test statistics. Findings The study found that combining particle swarm optimisation with back propagation in global and local search for attribute weights enhances the predictive accuracy of artificial neural networks. This also enhances transparency of the process, because it shows relative importance of attributes. Research limitations/implications A robust assessment of the models’ predictive accuracy was inhibited by fewer accuracy test statistics found in the software. The research demonstrates the efficacy of combining two models in the assessment of property values. Originality/value This work demonstrated the practicability of combining particle swarm optimisation with back propagation algorithms in finding optimal weights and training of the artificial neural networks within the mass appraisal environment.

The Hedonic Modelling of Rental Values of Properties in Nasarawa, Nigeria

“…Table 4 explains 68% variations in rental values, while Table 5 explains 64% variations in rental values. Unlike many empirical analysis that reveal the benefit of log transformation of the dependent variable with an enhancement in goodness of fit [18,25], the result in this study reveals otherwise. The quality of data used in this assessment would not permit a categorical statement on the best functional form specification for the local market.…”

“…Mass appraisal is the assessment of market values on a number of properties over a given time period using standardised techniques [17,18]. These techniques, especially the OLS, have been integrated as a part of CAMA systems by the property tax assessment community [19].…”

“…McCluskey [26] used 40,138 apartment sales in Astana, Kazakhstan and found the linear model to provide a good fit than the semi-log. Yacim and Boshoff [18] utilised the linear, semi-log and log-log specification on 3242 observations in Cape Town, South Africa and found the semi-log as the most preferred regression based model.…”

Mass Appraisal Models of Real Estate in the 21st Century: A Systematic Literature Review

“…The ANN performs well for modeling the non-linear relationship because of its characteristics of semi-parametric regression. In addition to the basic MRA, although researchers have to face the "black box" of the ANN's structure, it is still the most popular model used in AI-based models [36][37][38][39][40][41][42][43][44][45][46][47][48][49].…”

Comparative Performance of Machine Learning Classifiers in Detecting Performing and Non-performing Residential Property Renters

“…ML has continued to play crucial roles in many sectors due to its efficient cost-saving and less human psychological bias (Mutupe et al, 2017). The research focusing on ML in the real estate market is still very limited or sometimes not in existence especially in developing countries (Olawale, 2011 andYacim andBoshoff, 2017). Aside from this, choosing a particular classifier as being uniformly best in performance may be a bit difficult given the strengths and weaknesses of each of the methods, thus, searching for the best predictive model/classifier is still in progress .…”