Abstract
To support senior citizens’ smooth conversations, we have developed a conversation support system that provides several topics for speakers. To effectively support conversation, the system should select suitable topics and speaking styles for the speakers while observing topic-shift situations regarding their previous conversations. In this study, we discuss the differences in conversation styles and topic-shift characteristics between elderlies and those between youths. A comparison of conversations of elderly people and youths revealed two distinct characteristics. The casual conversations of elderlies tended to include long utterances with the speakers firmly expressing their knowledge or opinion. By contrast, the conversations of youths did not reveal long utterances and they tended not to express their opinions at once, but only gradually, after watching the reaction of the other person. Further, elderly speakers tended to talk on a topic more in-depth and did not get sidetracked compared with young speakers. Moreover, the boundary of topics in the case of elderly people tended to be clearer than in the case of youths. As a result of generation identification experiments using “the number of utterances by each topic,” “the number of content words in each utterance,” and “the deviation of word similarity,” the precision rate for youths is 90.5% but that for elderlies is 42.2%. This suggests that these parameters are effective to extracting the young speakers with certainty.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Arguello, J., et al.: Topic segmentation of dialogue. In: HLT-NAACL Workshop on Analyzing Conversations in Text and Speech, pp. 42–49 (2006)
Afantenos, S., et al.: Discourse parsing for multi-party chat dialogues. In: EMNLP, pp. 928–937 (2015)
Ryuichi, T., et al.: A weakly supervised method for topic segmentation and labeling in goal-oriented dialogues via reinforcement learning. In: 27th IJCAI-18, pp. 4403–4410 (2018)
Bai, P.S., et al.: Topic shifts in conversations: Focus on Malaysian Chinese teenagers. SARJANA, 26(2), 101–118 (2011)
Wu, Z., Palmer, M.: Verbs semantics and lexical selection. In: Proceedings of the 32nd Annual Meeting on ACL, pp. 133–138 (1994)
Batet, M., Sánchez, D.: A review on semantic similarity. In: Encyclopedia of Information Science and Technology Third Edition, vol. 10, pp. 7575–7583 (2015)
Acknowledgment
This work is supported by JSPS KAKENHI Grant Number 19K04934.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Iida, Y., Wakita, Y. (2021). Topic-Shift Characteristics of Japanese Casual Conversations Between Elderlies and Between Youths. In: Gao, Q., Zhou, J. (eds) Human Aspects of IT for the Aged Population. Supporting Everyday Life Activities. HCII 2021. Lecture Notes in Computer Science(), vol 12787. Springer, Cham. https://doi.org/10.1007/978-3-030-78111-8_28
Download citation
DOI: https://doi.org/10.1007/978-3-030-78111-8_28
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-78110-1
Online ISBN: 978-3-030-78111-8
eBook Packages: Computer ScienceComputer Science (R0)