Automated Scoring of Clinical Expressive Language Evaluation Tasks.
Automated Scoring of Clinical Expressive Language Evaluation Tasks.
复制标题
DOI：
10.18653/v1/2020.bea-1.18
复制
发表时间：
2020-07
期刊：
影响因子：
--
通讯作者：
Dolata J
中科院分区：
文献类型：
--
作者：
Wang Y;Prud'hommeaux E;Asgari M;Dolata J
许多用于诊断儿童语言障碍的临床评估工具包括一项任务，其中受试者必须使用特定的目标词来表达一个句子来描述一个图像。由于以这种方式产生句子需要说话者以复杂的方式整合句法和语义知识，因此通常会在几个不同的适当性维度上对响应进行评估，从而为每个响应产生单个综合得分。在本文中，我们提出了一个数据集组成的非临床引起的反应，三个相关的句子制定任务，我们提出了一种方法，自动评估其适当性。使用神经机器翻译，我们生成正确-不正确的句子对作为合成数据，以增加评分模型的训练数据的数量和多样性。我们的评分模型使用迁移学习来促进自动句子适当性评估。我们进一步将自定义单词嵌入与预训练的上下文嵌入进行比较，作为我们评分模型的特征。我们发现，迁移学习提高了评分的准确性，特别是在使用预先训练的上下文嵌入时。
Many clinical assessment instruments used to diagnose language impairments in children include a task in which the subject must formulate a sentence to describe an image using a specific target word. Because producing sentences in this way requires the speaker to integrate syntactic and semantic knowledge in a complex manner, responses are typically evaluated on several different dimensions of appropriateness yielding a single composite score for each response. In this paper, we present a dataset consisting of non-clinically elicited responses for three related sentence formulation tasks, and we propose an approach for automatically evaluating their appropriateness. Using neural machine translation, we generate correct-incorrect sentence pairs to serve as synthetic data in order to increase the amount and diversity of training data for our scoring model. Our scoring model uses transfer learning to facilitate automatic sentence appropriateness evaluation. We further compare custom word embeddings with pre-trained contextualized embeddings serving as features for our scoring model. We find that transfer learning improves scoring accuracy, particularly when using pre-trained contextualized embeddings.
登录
查看更多内容
影响因子：
5.4
作者：
Barnhoorn JS;Haasnoot E;Bocanegra BR;van Steenbergen H
通讯作者：
van Steenbergen H
DOI：
10.1111/j.1469-7610.2005.01584.x
发表时间：
2006-06-01
影响因子：
7.6
作者：
Conti-Ramsden, Gina;Simkin, Zoe;Botting, Nicola
通讯作者：
Botting, Nicola
影响因子：
7.6
作者：
Norbury, Courtenay Frazier;Gooch, Debbie;Pickles, Andrew
通讯作者：
Pickles, Andrew
影响因子：
2.8
作者：
Grimm, Angela;Schulz, Petra
通讯作者：
Schulz, Petra