A Procedure for Extending Input Selection Algorithms to Low Quality Data in Modelling Problems with Application to the Automatic Grading of Uploaded Assignments

<table>Supports (dashed lines), modal points (bars), fuzzy ranks (abscissa), and crisp ranks (ordinate of the squares) of the 50 most relevant metrics. Those metrics whose square is plotted below the diagonal line occupy a more relevant position under the fuzzy rank.</table>

The Scientific World Journal

fig2

Figure 2

Figure 2: A Procedure for Extending Input Selection Algorithms to Low Quality Data in Modelling Problems with Application to the Automatic Grading of Uploaded Assignments 

Figure 2 | A Procedure for Extending Input Selection Algorithms to Low Quality Data in Modelling Problems with Application to the Automatic Grading of Uploaded Assignments 