University of California, Santa Cruz - SlugBot

Kevin B.

- Team Leader

Kevin is a first year PhD student working under Marilyn Walker in the Natural Language and Dialogue Systems lab since the final year of his undergraduate studies. Currently his research is focused on task-oriented dialogue systems and expressive chat bots. His previous work includes representing monologic content as dialogic retellings, depicting personality in dialogue, and sentiment classification of reviews. Recently he participated in the eNTERFACE’16 workshop where he was responsible for the dialogue management and WoZ components of a virtual journalist agent modeled with the Aria Valuspa toolkit.

Amita M.

Amita is a fifth year PhD student working with Marilyn Walker at the Natural Language and Dialogue Systems lab. Her research interests span the field of natural language processing and statistical machine learning. Her current work involves text mining, opinion extraction from text, sentiment analysis with a special focus on understanding opinions and arguments in social media. She works on large scale projects involving stance and sentiment classification in conversational data from Twitter, debate forums and has extensive experience with NLP tools such as parsers, dictionaries and deep learning frameworks.

Jiaqi W.

Jiaqi is a third year PhD student. Her supervisor is Marilyn Walker at the Natural Language and Dialogue Systems lab. Her current work focus on sentiment analysis and event extraction based on the Stanford CoreNLP parser and SAMEFOR FrameNet parser.

Shereen O.

Shereen is a third year PhD student working with Marilyn Walker at the Natural Language and Dialogue Systems lab. Her current research focuses on sarcasm classification in social media dialogue. She has also worked on sentiment analysis of personal blogs and Arabic product reviews, and characterizing argumentation styles in debate forums. She recently completed an internship at IBM Research, where she worked on sequential modeling of conversations in social media.

Marilyn Walker

- Faculty Advisor

Marilyn's research group at UCSC focuses on statistical and expressive natural language generation and dialogue models and systems. She developed the first dialog system "ELVIS", whose dialog manager was trained using reinforcement learning with feedback from human interaction, rather than via user simulation. She has over 15 years experience developing statistical generation algorithms for dialogue, including methods that both adapt to the user and express the system's personality, and thirty years of experience developing dialogue systems. She has worked in industry as well as academia, including developing dialogue systems and evaluation methods for dialogue at AT&T Labs Research.

Meet the 2017 Teams

Alexa Prize Socialbot Grand Challenge 3