Green building valuation based on machine learning algorithms / Thuraiya Mohd ... [et al.]

In the cycle of Industrial Revolution 4.0 (IR 4.0), many issues in the industries can be solved with implementation of artificial intelligence approaches, including machine learning models. Designing an effective machine learning model for prediction and classification problems is a continuous effort. In addition, time and expertise are important factors needed to adapt the model to a specific problem such as green building housing development. Green building is known as a potential method to improve building performance efficiency. To our knowledge, there is still no implementation of machine learning models on green building valuation features for building price prediction compared to conventional building development. This paper provides an empirical study report, that building price predictions are based on green building and other general determinants. This experiment used five common machine learning algorithms namely 1) Linear Regressor, 2) Decision Tree Regressor, 3) Random Forest Regressor, 4) Ridge Regressor and 5) Lasso Regressor tested on a real estate data-set of covering Kuala Lumpur District, Malaysia. 3 set of experiments was conducted based on the different feature selections and purposes The results show that the implementation of 16 variables based on Experiment 2 has given a promising effect on the model compare the other experiment, and the Random Forest Regressor by using the Split approach for training and validating data-set outperformed other algorithms compared to Cross-Validation approach. The research will provide an appropriate model in predicting the price of a green building which is beneficial to the government agencies and industry practices