Proposed agorithm for regression-based prediction with bulk noise

Chanintorn Jittawiriyanukoon


The noise has incited an original data due to a network with an inferior SNR. In case of the bulk noise, the insightful content within the data is substantially squeezed.  A cost-effective method will challenge to quarantine the insights, so that information can be utilized more resourcefully.  To achieve this aim, it is essential to iron the bulk noise content out, and then calculate the analytics of the clean data. As noise is bulk so some statistical methodologies such as averaging or randomizing are employed. A prediction using the regression-based model with bulk noise for the experiment in practice is introduced. The decomposition approach to separate the insights is exploited. The proposed algorithm achieves a (local) solution at each computing step and selects the best solution in view of global impacts. The correlation coefficient, average error, absolute error and mean squared error are used to constitute the prediction. Results from MOA simulation will be compared to actual data in the succeeding time. The prediction with bulk noise using the proposed algorithm outperforms.


Bulk noise; Mean Absolute Error; Mean Square Error; Noisy and missing data; Regression-based prediction; Signal-to-Noise Ratio;

Full Text:


Total views : 87 times


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

shopify stats IJEECS visitor statistics