Research Article

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

Figure 4

(a) Boxplot showing the statistical differences between the test error of the combination of Neural Networks (NN), Support Vector Machines (SVM), Regression Trees (RT), Random Forests (RF), and NMIC with the feature set computed as described in this paper. NMIC exploits the imprecision in the information better than the alternatives. (b) Membership functions of the ranks of the first (solid) and 10th (dashed) features, that is, COCOMO SLOC and McCabe complexity.
468405.fig.004a
(a)
468405.fig.004b
(b)