Topic Shift Detection for Mixed Initiative Response

Rachna Konigari, Saurabh Ramola, Vijay Vardhan Alluri, Manish Shrivastava


Abstract
Topic diversion occurs frequently with engaging open-domain dialogue systems like virtual assistants. The balance between staying on topic and rectifying the topic drift is important for a good collaborative system. In this paper, we present a model which uses a fine-tuned XLNet-base to classify the utterances pertaining to the major topic of conversation and those which are not, with a precision of 84%. We propose a preliminary study, classifying utterances into major, minor and off-topics, which further extends into a system initiative for diversion rectification. A case study was conducted where a system initiative is emulated as a response to the user going off-topic, mimicking a common occurrence of mixed initiative present in natural human-human conversation. This task of classifying utterances into those which belong to the major theme or not, would also help us in identification of relevant sentences for tasks like dialogue summarization and information extraction from conversations.
Anthology ID:
2021.sigdial-1.17
Volume:
Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue
Month:
July
Year:
2021
Address:
Singapore and Online
Venue:
SIGDIAL
SIG:
SIGDIAL
Publisher:
Association for Computational Linguistics
Note:
Pages:
161–166
Language:
URL:
https://aclanthology.org/2021.sigdial-1.17
DOI:
Bibkey:
Cite (ACL):
Rachna Konigari, Saurabh Ramola, Vijay Vardhan Alluri, and Manish Shrivastava. 2021. Topic Shift Detection for Mixed Initiative Response. In Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 161–166, Singapore and Online. Association for Computational Linguistics.
Cite (Informal):
Topic Shift Detection for Mixed Initiative Response (Konigari et al., SIGDIAL 2021)
Copy Citation:
PDF:
https://preview.aclanthology.org/ingestion-script-update/2021.sigdial-1.17.pdf
Video:
 https://www.youtube.com/watch?v=0QAMrBoEwmk