Figure (4). Results from KNN with all features, the green dots is the ratio of actual value of site EUI over predicted value, and the blue line is to show that the more the dots be closer to it, the prediction is more accurate.
5.3 Support Vector Machine
5.3.1 Support Vector Machine Regression
The results from SVM are not very promising in terms of the accuracy score (R2). With all features included in the mode, the optimized parameters for the residential dataset are: epsilon: 0.0, C: 1, gamma: 0.0001, from this model we have result of test score R2 as -0.006 and mean squared error as 841.76. For commercial buildings, the optimized parameters for the commercial dataset are: epsilon: 0.0, C: 501, gamma: 0.01, and the R2 from this model is -0.004 along with 1181.74 of mean squared error. Results shown as below in Figure (5).