[“Explainability for Natural Language Processing”]

This tutorial is part of the Explainable AI Research Project

Slides

Program

Presenters

Marina Danilevsky is a Research Staff Member at IBM Research - Almaden whose work centers on understanding structured and unstructured text data. An important research direction for her has been in working on specific domains such as legal, financial, and communications, with a particular focus on model explainability and human-in-the-loop techniques. She received her PhD in Computer Science from the University of Illinois atUrbana-Champaign
Shipi Dhanorkar is a PhD student in Informatics fo-cusing on Human Centered Design at Pennsylvania State Univer-sity where she is also pursuing a graduate minor in Social DataAnalytics. Her research has explored community engagement andcomputer supported collaborative work. She is interested in human-AI interactions and research that promotes meaningful human-AIcollaborations. She holds a BE in Computer Engineering and an MAin Educational Policy. This work is done during her summer intern-ship at IBM Research - Almaden.
Yunyao Li is a Distinguished Research Staff Memberand Senior Research Manager with IBM Research - Almaden, whereshe manages the Scalable Knowledge Intelligence Department. Sheis a Master Inventor and a member of the IBM Academy of Technology. She is also a member of the New Voices program of the American National Academies. Her expertise is in the interdisciplinary areas of natural language processing, databases, human-computer interaction, and information retrieval. She is interested in designing, developing, and analyzing large-scale systems that are usableby a wide spectrum of users. Her current research focuses on scalable natural language processing and knowledge intelligence. She regularly gives talks and tutorials at conferences and universities across the globe, including one on “Transparent machine learningfor information extraction” in EMNLP 2015. She received her Ph.D. from the University of Michigan, Ann Arbor.
Lucian Popa is a Principal Research Staff Member and Manager at IBM Research - Almaden. He is known for work on dataexchange and schema mapping, for which he received two Test-of-Time Awards in ICDT 2013 and PODS 2014, and for work on human-in-the-loop tools and foundations for entity resolution, for which hereceived a Best Paper Award in ICDT 2015. He is also a winner of the 2020 Alonzo Church Award for Outstanding Contributions to Logic and Computation. Lucian has contributed to several IBM products in the area of information integration and entity resolution, and heis an ACM Distinguished Member.
Kun Qian is an Applied Scientist in the Search Scienceand AI group at Amazon. Prior to joining Amazon, he was a Research Staff Member at IBM Research - Almaden. He is broadly interested in Artificial Intelligence, Natural Language Processing,Data Integration, and Search. He has been working on designing and building human-in-the-loop machine learning systems for several entity-centric problems including entity resolution and entity normalization. His research has been published in notable AI/DB/HCI conferences such as AAAI, ACL, EMNLP, COLING, VLDB, PODS, ICDE, TODS, CIKM, IUI, ACM DIS, ISWC, etc.
Anbang Xu is a Manager and Research Staff Memberin the USER group at IBM Research - Almaden. He manages a team that leads data-driven efforts to train and enhance Watson AI services through the development and integration of techniques in Crowdsourcing, Machine Learning, and Visualization. His work is a mix of HCI and AI. His recent focus is on the design of chatbots and personality analytics systems for user engagement on social media. He has been speaker and panelist at numerous conferences in HCI and Applied Machine Learning.

If you find this tutorial useful, please cite our work:

@inproceedings{danilevsky2021explainability,
  title={Explainability for Natural Language Processing},
  author={Danilevsky, Marina and Dhanorkar, Shipi and Li, Yunyao and Popa, Lucian and Qian, Kun and Xu, Anbang},
  booktitle={Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery \& Data Mining},
  pages={4033--4034},
  year={2021}
}