Consequently, multiple predictions for each compound should be combined as follows: For numeric predictions, the mean predicted value is preferable.
In automatic ARIMA modeling-based data aggregation scheme, ordinary sensor node will transmit the sensed data value to the aggregator only when the prediction error between sensed value and predicted value is beyond the application-specified error threshold.
When the prediction error between the sensed value and predicted value is within the application-specified error threshold, sensor node will not transmit the sensed value to the aggregator.
In particular this means predictions from the several machines are not averaged to make the prediction on the test case, but rather the predicted value is a weighted average of the original outcomes.
In the classical PE embedding method proposed by Thodi and Rodriguez, the predicted value is rounded to integer number for integer prediction-error expansion (IPE) embedding.
Then, the accuracy of the predicted value is reported based on the accuracy of prediction of the neighboring chemicals in the training set using a leave-one-out procedure, weighted by similarity to the query chemical.
