Explainable AI for Natural Language Processing
Presented at the 1st Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 10th International Joint Conference on Natural Language Processing
Shipi Dhanorkar, Christine Wolf, Kun Qian, Anbang Xu, Lucian Popa and Yunyao Li
Led a cutting-edge tutorial on the issues of transparency and interpretability of AI models. Both the research community and industry have been developing new techniques to render black-box Natural Language Processing (NLP) models more transparent and interpretable. This tutorial was based on my internship work in which I presented along with an interdisciplinary team comprising of human-computer interaction (HCI), and NLP researchers. The tutorial provided an introduction to explainable AI (XAI) and a review of the state-of-the-art for explainability research in NLP; and findings from a qualitative interview study with data scientists, AI engineers, technical strategists, and UX designers identifying practical challenges and AI model explainability concerns that arise in real-world deployments. http://www.aacl2020.org/program/tutorials/
Crowdsourcing Change: A novel vantage point for investigating online petitioning platforms.
In International Conference on Information, Springer, iConference, 2019
Dhanorkar, S. and Rosson, M.B.
Comparative analysis of design features of four online petitioning platforms -Change.Org, Care2 Petitions, We the People and MoveOn.Org. Applied crowdsourcing platforms’ design principles and developed design implications for (i) incentives/ motivations to participate, (ii)reputation mechanisms, (iii)critical mass for collective action and (iv)message framing.
Can we nudge users toward better password management? An initial study
In Proceedings of ACM CHI Conference Extended Abstracts, 2018
Dhanorkar(Kankane), S., DiRusso, C., and Buckley, C.
Online between-subjects experiment to study effectiveness of different nudges [link to paper]
Strengthening community data: Towards pervasive participation.
In Proceedings of Digital Government Research: Governance in the Data Age, dg.o 2018
Carroll, J. M., Beck, J., Dhanorkar, S., Binda, J., Gupta, S., & Zhu, H.
Proposed the concepts of 'community data' and its role in community development, learning, decision making and governance [link to paper].