当前位置:
X-MOL 学术
›
J. Assoc. Inf. Sci. Technol.
›
论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A taxonomy, data set, and benchmark for detecting and classifying malevolent dialogue responses
Journal of the Association for Information Science and Technology ( IF 3.5 ) Pub Date : 2021-05-26 , DOI: 10.1002/asi.24496 Yangjun Zhang 1 , Pengjie Ren 2 , Maarten de Rijke 1, 3
Journal of the Association for Information Science and Technology ( IF 3.5 ) Pub Date : 2021-05-26 , DOI: 10.1002/asi.24496 Yangjun Zhang 1 , Pengjie Ren 2 , Maarten de Rijke 1, 3
Affiliation
Conversational interfaces are increasingly popular as a way of connecting people to information. With the increased generative capacity of corpus-based conversational agents comes the need to classify and filter out malevolent responses that are inappropriate in terms of content and dialogue acts. Previous studies on the topic of detecting and classifying inappropriate content are mostly focused on a specific category of malevolence or on single sentences instead of an entire dialogue. We make three contributions to advance research on the malevolent dialogue response detection and classification (MDRDC) task. First, we define the task and present a hierarchical malevolent dialogue taxonomy. Second, we create a labeled multiturn dialogue data set and formulate the MDRDC task as a hierarchical classification task. Last, we apply state-of-the-art text classification methods to the MDRDC task, and report on experiments aimed at assessing the performance of these approaches.
中文翻译:
用于检测和分类恶意对话响应的分类法、数据集和基准
对话式界面作为一种将人们与信息联系起来的方式越来越受欢迎。随着基于语料库的对话代理生成能力的增加,需要对内容和对话行为方面不适当的恶意响应进行分类和过滤。以前关于检测和分类不当内容的研究主要集中在特定类别的恶意或单个句子而不是整个对话。我们为推进恶意对话响应检测和分类 (MDRDC) 任务的研究做出了三项贡献。首先,我们定义任务并提出一个分层的恶意对话分类法。其次,我们创建了一个带标签的多轮对话数据集,并将 MDRDC 任务制定为分层分类任务。最后的,
更新日期:2021-05-26
中文翻译:
用于检测和分类恶意对话响应的分类法、数据集和基准
对话式界面作为一种将人们与信息联系起来的方式越来越受欢迎。随着基于语料库的对话代理生成能力的增加,需要对内容和对话行为方面不适当的恶意响应进行分类和过滤。以前关于检测和分类不当内容的研究主要集中在特定类别的恶意或单个句子而不是整个对话。我们为推进恶意对话响应检测和分类 (MDRDC) 任务的研究做出了三项贡献。首先,我们定义任务并提出一个分层的恶意对话分类法。其次,我们创建了一个带标签的多轮对话数据集,并将 MDRDC 任务制定为分层分类任务。最后的,