An Empirical Study for Adopting Machine Learning Approaches for Gas Pipeline Flow Prediction

Figure 10

Performance variation gives different neighbor numbers and averaging methods. The results suggest that, when taking the distance between the given data point and neighbors as weight, the performance gets better as the number of neighbors increases. However, the performance gets worse as the number of neighbors increase when using equal weights for each neighbor. These results suggest that weight based on distance is a better choice for this problem. (a) MSE, (b) MRE, (c) .

(a)

(b)

(c)