A Learning-based Sampling Approach to Extractive Summarization

In this paper we present a novel resampling model for extractive meeting summarization. With resampling based on the output of a base-line classifier, our method outperforms previ-ous research in the field. Further, we com-pare an existing resampling technique with our model. We report on an extensive se-ries of experiments on a large meeting corpus which leads to classification improvement in weighted precision and f-score.