Confidence Measures for Error Discrimination in an Interactive
Predictive Parsing Framework
We study the use of Confidence Measures (CM) for erroneous constituent discrimination in an Interactive Predictive Parsing (IPP) framework. The IPP framework al- lows to build interactive tree annotation systems that can help human correctors in constructing error-free parse trees with little effort (compared to manually post-editing the trees obtained from an auto- matic parser). We show that CMs can help in detecting erroneous constituents more quickly through all the IPP process. We present two methods for precalculat- ing the confidence threshold (globally and per-interaction), and observe that CMs remain highly discriminant as the IPP process advances.