What Do Users Care About? Detecting Actionable Insights from User Feedback

Kasturi Bhattacharjee, Rashmi Gangadharaiah, Kathleen McKeown, Dan Roth


Abstract
Users often leave feedback on a myriad of aspects of a product which, if leveraged successfully, can help yield useful insights that can lead to further improvements down the line. Detecting actionable insights can be challenging owing to large amounts of data as well as the absence of labels in real-world scenarios. In this work, we present an aggregation and graph-based ranking strategy for unsupervised detection of these insights from real-world, noisy, user-generated feedback. Our proposed approach significantly outperforms strong baselines on two real-world user feedback datasets and one academic dataset.
Anthology ID:
2022.naacl-industry.27
Volume:
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track
Month:
July
Year:
2022
Address:
Hybrid: Seattle, Washington + Online
Editors:
Anastassia Loukina, Rashmi Gangadharaiah, Bonan Min
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
239–246
Language:
URL:
https://aclanthology.org/2022.naacl-industry.27
DOI:
10.18653/v1/2022.naacl-industry.27
Bibkey:
Cite (ACL):
Kasturi Bhattacharjee, Rashmi Gangadharaiah, Kathleen McKeown, and Dan Roth. 2022. What Do Users Care About? Detecting Actionable Insights from User Feedback. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track, pages 239–246, Hybrid: Seattle, Washington + Online. Association for Computational Linguistics.
Cite (Informal):
What Do Users Care About? Detecting Actionable Insights from User Feedback (Bhattacharjee et al., NAACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.naacl-industry.27.pdf
Video:
 https://aclanthology.org/2022.naacl-industry.27.mp4
Data
CLINC150