Are All the Datasets in Benchmark Necessary? A Pilot Study of Dataset Evaluation for Text Classification

Anonymous

Abstract: In this paper, we ask the research question if all the datasets in the benchmark are necessary. We approach this by first characterizing the distinguishing ability of datasets when comparing different systems. Experiments on 9 datasets and 36 systems show that several existing benchmark datasets contribute little to discriminating top-scoring systems, while those less used datasets exhibit impressive discriminative power. We further, taking the text classification task as a case study, investigate the possibility of predicting dataset discrimination based on its properties (e.g., average sentence length). Our preliminary experiments promisingly show that given a sufficient number of training experimental records, a meaningful predictor can be learned to estimate dataset discrimination over unseen datasets. We released all related code at Github \url{https://github.com/annonnlp-demo/acl-V2} and a new benchmark dataset for text classification based on our observations.

Paper Type: long