Some effects of database corruption in system prediction performance
Year: 2011
Type: conference-paper
Abstract: Many types of intelligent adaptive systems use\nvast databases of a-priori knowledge during training\nphases. These systems are then reliant on both the accuracy\nof this data and on the breadth of the data. It is assumed\nwhilst training that the data encompasses the total\noperating window for the system in enough detail to\ngenerate an accurate ‘black box’ model of the plant under\ncontrol. It may be that under certain unforeseen operating\nconditions, or in a scenario where there is little prior\nknowledge, the system may be forced to operate outside the\nscope of the original a-priori knowledge. Lastly the data\ngathered into the a-priori source may have been\nunintentionally corrupted. This paper aims to examine some\nof these effects upon two common adaptive intelligent tools,\nneural networks and an adaptive neuro-fuzzy inference\nsystem, ANFIS, network.
Authors: Matthew R. Chamberlain, Michael R. Jackson, Robert Parkin
Institution: Loughborough University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W2755650816