Several factors, such as neighborhood, garage availability, assessment class, distances from center, public transport and green spaces were found to be significant predictors of property values in Edmonton. This implies that these variables are crucial elements that must to be taken into account when performing future property assessments.
I split the dataset into training and test set with a ratio of 85 to 15 and then made the prediction on the test set using the training set. I tried different ratios for all but this was the best ratio in terms of the metrics and the prediction I wanted from this data.
According to the results of the research, neural networks performed best among all the models used as it gave the best result for test set error rate, RMSE (Root Mean Squared Error), MAE (Mean Absolute Error) which were the metrics best used for the model selection. This shows that using machine learning approaches to enhance the precision of property value estimates may be beneficial in the future.
Also, by doing feature selection we came to know what are the major variables responsible for property assessment values.
I split the dataset into training and test set with a ratio of 85 to 15 and then made the prediction on the test set using the training set. I tried different ratios for all but this was the best ratio in terms of the metrics and the prediction I wanted from this data.
According to the results of the research, neural networks performed best among all the models used as it gave the best result for test set error rate, RMSE (Root Mean Squared Error), MAE (Mean Absolute Error) which were the metrics best used for the model selection. This shows that using machine learning approaches to enhance the precision of property value estimates may be beneficial in the future.
Also, by doing feature selection we came to know what are the major variables responsible for property assessment values.
Disclaimer: This website's study design, data analysis, and subsequent results, discussion, and conclusions were created as part of a project for University of Alberta's Ren R 690 course. The assignment's parameters should not be construed outside of them because all data and findings are regarded as preliminary.