A Semieager Classifier for Open Relation Extraction

Figure 1

The SemiE. A set of positive and negative training examples is shown, along with a query instance to be classified. The query instance is assigned to the positive examples on the left, whereas the algorithm classifies it as negative on the right.

(a) The number of positive examples is identical to that of negative examples. And the query instance is near the positive center

(b) The positive examples are less than the negative examples. But the query instance has the same distance to positive and negative centers