Dynamic soft sensors with active forward-update learning for selection of useful data from historical big database

Stream-based active learning with linear models

“…Regression models, on the other hand, are extremely useful in the development of soft sensors for hard-to-measure process variables or in quality control problems where a product's characteristic is measured on a continuous scale. That is why active learning in conjunction with regression models is capturing the interest of many researchers [7][8][9][10].…”

Active Learning Strategy for Online Prediction of Particle Size Distribution in Cobalt Oxalate Synthesis Process

“…The prediction accuracy of the two methods can be reflected by the root mean square error (RMSE). For the testing dataset, RMSE is given as follows (28) where k = 1, 2, . .…”

“…In this work, we assume that a massive amount of process data samples while only a very limited number of labeled samples are obtained. The main objective of the active learning method is to opportunely label the most useful unlabeled data from the training dataset to maximize the estimation performance of LSSVR model while minimizing the number of samples for modeling [28]. Therefore how to construct an index to evaluate the informativeness of the data is the main issue for the active learning strategy.…”

Review on Fault Detection and Diagnosis Feature Engineering in Building Heating, Ventilation, Air Conditioning and Refrigeration Systems

“…This is because, in some practical cases, it is difficult to obtain sufficient fault data and labels. Moreover, active learning [295][296][297][298] and transfer learning [230,299] which can address the issues of real-life fault detection and diagnosis cases using unlabeled data should be seriously considered. However, the negative transfer should be avoided in engineering scenarios.…”