A Procedure for Extending Input Selection Algorithms to Low Quality Data in Modelling Problems with Application to the Automatic Grading of Uploaded Assignments
Figure 1
Fuzzy representation of LQD. (a) 1000 instances of an interval-valued database. (b) Fuzzy (possibilistic) representation of the mean value of the data in the left part. -cuts of this fuzzy set are bootstrap-based confidence intervals of the interval valued data; that is, they are the smallest intervals containing at least a fraction of the data.