Building a Discourse Parser | o S: A binary classifier , for structure (existence of a connecting node between the two input sub-trees). |
Building a Discourse Parser | Because the original SVM algorithms build binary classifiers , multi-label classification requires some adaptation. |
Building a Discourse Parser | A possible approach is to reduce the multi-classification problem through a set of binary classifiers , each trained either on a single class (“one vs. all”) or by pair (“one vs. one”). |
Evaluation | Binary classifier S is trained on 52,683 instances (split approximately 1/3, 2 / 3 between positive and negative examples), extracted from 350 documents, and tested on 8,558 instances extracted from 50 documents. |
Baseline Approaches | More specifically, both baselines recast the cause identification problem as a set of 14 binary classification problems, one for predicting each shaper. |
Baseline Approaches | In the binary classification problem for predicting shaper 3,, we create one training instance from each document in the training set, labeling the instance as positive if the document has 3,- as one of its labels, and negative otherwise. |
Baseline Approaches | After creating training instances, we train a binary classifier , Ci, for predicting 3i, employing as features the top 50 unigrams that are selected according to information gain computed over the training data (see Yang and Pedersen (1997)). |
Introduction | Second, the fact that this is a 14-class classification problem makes it more challenging than a binary classification problem. |