Pearl Pu is a Chinese-born Swiss computer scientist born in Shanghai, China, known for her work on human-computer interaction with Artificial Intelligence systems, starting with work on using case-based reasoning for intelligent Computer-aided design. Her research focuses on recommender systems, conversational AI, human-computer interaction, artificial intelligence, and the ethics of artificial intelligence. Associated with the University of Pennsylvania, Norman Badler, and the École Polytechnique Fédérale de Lausanne (EPFL), her lab is located at https://hci.epfl.ch.
Career
After completing her Ph.D. from the University of Pennsylvania, she obtained the Research Initiation Award (known as the Career Award) from the NSF before moving to Switzerland. She has been a faculty member and senior scientist at École Polytechnique Fédérale de Lausanne, where she founded the Human-Computer Interaction Group in 2000.[2,1] In 1997, she co-founded the company Iconomic Systems SA, known for the development of an agent-based paradigm for travel e-commerce, and served as chairperson until its sale to i:FAO in 2001.[3] She spent six months each as a visiting scholar at Stanford University in 2001 and at the Hong Kong University of Science and Technology in 2010.
Pu has developed methods for obtaining accurate user preferences through dialogue-based recommender systems.[4] She has served on editorial boards and conference committees such as the International Joint Conference on Artificial Intelligence, AAAI Conference on Artificial Intelligence, The Web Conference, and Conference on Human Factors in Computing Systems. She was also general or program chair of the ACM conferences on Electronic Commerce, Recommender Systems, and Adaptive Hypermedia.[8,5,6]
Honors
In 2014, Pu was the winner of the French "2030 World Innovation Challenge" for the Livelyplanet project.[9,10] Her technology found particular echo in the French press.[11] Pu was made a EurAI fellow in 2021 and named a distinguished speaker of the Association for Computing Machinery.[12]
Selected publications
Pu's most cited work is on the ResQue framework for evaluating recommender systems, represented by "A user-centric evaluation framework for recommender systems" (with Li Chen and Rong Hu), published in the Proceedings of the fifth ACM Conference on Recommender Systems (pp. 157–164, 2011).
Representative of her work in intelligent computer-aided design and case-based reasoning are "Issues and Applications of Case-Based Reasoning to Design" (with Mary-Lou Maher), published by Psychology Press in 1997 (ISBN 978-0805823134), and "COMPOSER: A case-based reasoning system for engineering design" (with Lisa Purvis) in Robotica (16(3), pp. 285–295, 1998).
Her research on humans interacting with artificial intelligence through explanation and visualization is represented by "Visualizing Resource Allocation Tasks" (with George Melissargos) in IEEE Computer Graphics and Applications (IEEE Computer Society Press, 17(4), pp. 6–9, 1997) and "Trust-inspiring explanation interfaces for recommender systems" (with Li Chen) in the journal Knowledge-Based Systems (20(6), pp. 542–556, 2007).
Representative of her work on interactive recommender systems and example-critiquing are "Preference-based search using example-critiquing with suggestions" (with Paolo Viappiani and Boi Faltings) in the Journal of Artificial Intelligence Research (27, pp. 465–503, 2006) and "Critiquing-based recommenders: survey and emerging trends" (with Li Chen) in User Modeling and User-Adapted Interaction (22(1-2), pp. 125–150, 2012). Additionally, representative of her work on behavior recommender systems are "HealthyTogether: exploring social incentives for mobile fitness applications" (with Yu Chen) in the Proceedings of the Second International Symposium of Chinese CHI (pp. 25–34, 2014) and "Can Fitness Trackers Help Diabetic and Obese Users Make and Sustain Lifestyle Changes?" (with Yu Chen and Mirana Randriambelonoro) in IEEE COMPUTER (50(3), pp. 20–29, 2017).
