Application of intelligent recommender systems to data reconstruction and denoising
Uday Hegde, Zheng Peng Yuan, M. Bahadori · 39th Aerospace Sciences Meeting and Exhibit · 2001
The application of recommender systems to data reconstruction and denoising is explored in this paper by applying three separate techniques to oscillatory temperature data from a spaceflight combustion experiment. The techniques utilized are neural networks, collaborative filtering, and multiple linear regression. Issues explored include rapidity of convergence of system parameters, optimal and sub-optimal systems, and effect of varying statistics on system predictions.