The
Star Student Project classification
process is composed of two steps: learning from the training data set and
labeling the unclassified data. The training data set S consists of
n elements sj; j ¼ 1;
. . . ; n, characterized by D attributes a1; . . . ; aD. Additionally, one of
the K classes is assigned to each data element. Each attribute ai; i 2 f1; . .
.;Dg takes values from the corresponding set Ai.