MACHINE LEARNING IN CULTURAL HERITAGE PRICE PREDICTION : ALGORITHMS AND FACTORS OF HERITAGE
Video
Abstract
The aim of this study is to identify the best
model of machine learning algorithms in predicts the price of heritage datasets.
This study has proposed 114 units of prewar shophouses in George Town Penang
Island from 2004 until 2019 and applied five machine learning algorithms for
training heritage dataset by evaluating using R-Square and Root Mean Square
Error. The finding shows the consistency of the random forest model as the best
model as price prediction from previous studies including in this study which
has been selected as the best machine learning model for heritage price prediction
with the highest R-Square and lowest Root Mean Square Error. The use of machine
learning techniques as price prediction using heritage datasets are not actively
been discovered by other researchers, thus this study has extending previous
studies by restructuring the price prediction training configuration and update
factors of heritage. The authors have summarised the particular machine learning
algorithms and current factors of heritage by analyzing previous literature
and publication from George Town World Heritage Incorporated. This finding
broadly supports the work of other studies in this area which linking the price
and factor of contribution. Besides that, this study has brought to light the potential
implications of misuse or misunderstanding of heritage to be interpreted incorrect context and suggestions for future research works.
Team Member
- NUR SHAHIRAH BINTI JA'AFAR
- JUNAINAH MOHAMAD
Result
