Topic Segmentation for Interview Dialogue System
Topic Segmentation for Interview Dialogue System
复制标题
DOI：
10.1145/3508230.3508237
复制
发表时间：
2021-12
期刊：
影响因子：
--
通讯作者：
Taiga Kirihara;Kazuyuki Matsumoto;M. Sasayama;Minoru Yoshida;K. Kita
中科院分区：
文献类型：
--
作者：
Taiga Kirihara;Kazuyuki Matsumoto;M. Sasayama;Minoru Yoshida;K. Kita
本研究利用访谈对话语料进行话题切分。在已有的访谈对话语料库中添加话语意图标签，并使用BERT、Sentence BERT和Distil BERT对话语进行矢量化。此外，主题分类进行使用的话语意图标签和前面和后面的说出的句子的特征。因此，当话语意图标签与DistilBERT一起使用时，实现了最大的准确性。
In this study, topic segmentation was performed by referring to the interview dialogue corpus. Utterance intention tags were added to the existing interview dialogue corpus, and uttered sentences were vectorized using BERT, Sentence BERT, and Distil BERT. In addition, topic classification was performed using the utterance intention tags and the features of the preceding and following uttered sentences. Consequently, the greatest accuracy was achieved when the utterance intention tag was used with DistilBERT.