D Powers - International Journal of Corpus Linguistics, 1997 - jbe-platform.com
… We conclude with an evaluation of the relative utility of a large array of different metrics and … is a report on a language learning project, our approach to evaluatingmetrics is in no way …
H Jia, Y Cheung, J Liu - … on neural networks and learning …, 2015 - ieeexplore.ieee.org
… a metric to quantify the distance between categorical data for unsupervisedlearning well. In this … The evaluations of clustering outcomes obtained by the k-modes algorithm with different …
… We investigate evaluationmetrics for dialogue … metrics from machine translation to compare a model’s generated response to a single target response. We show that these metrics …
RY Pang, K Gimpel - arXiv preprint arXiv:1810.11878, 2018 - arxiv.org
… 2018), but all either lack certain categories of unsupervisedmetric or lack human validation of them, which we contribute. Moreover, the textual transfer community lacks discussion of …
Y Gao, W Zhao, S Eger - arXiv preprint arXiv:2005.03724, 2020 - arxiv.org
… for evaluating multidocument summaries, we investigate unsupervisedevaluation methods, which … In particular, we focus on evaluating the relevance (Peyrard, 2019) of multi-document …
… As robustly capturing these statistical dependencies is a crucial step of the evaluation metrics that do not rely on interventions, we argue that future work on disentanglement scores …
J Belouadi, S Eger - arXiv preprint arXiv:2202.10062, 2022 - arxiv.org
… unsupervisedevaluationmetrics. To do so, we leverage similarities and synergies between evaluationmetric … In particular, we use an unsupervisedevaluationmetric to mine pseudo-…
… in unsupervisedlearning makes the question of evaluation and cluster quality assessment more complicated than in supervised learning. So … of the features and metrics used during the …
… the model parameters through Reinforcement Learning (RL) techniques. This makes the choice of a good evaluationmetric even more important. Unfortunately, ROUGE is known to …