The UCSB brown-bag forum on spatial thinking presents

Interactive Spatial Representations for Search and Recommendation Algorithms

John O’Donovan & Byungkyu (Jay) Kang, Computer Science, UCSB

12:00 p.m. Tuesday, November 17, 2015 | Phelps Hall 3512 (map)

Abstract. Search and recommendation algorithms have become standard tools that most of us interact with in our daily lives. Whether it is a movie from Netflix, an advert on Facebook, a result from a Google query or a product recommendation from Amazon, these complex algorithms work behind the scenes to passively or actively learn about user tastes and preferences. This information is used in turn to personally tailor the information space for the user. However, these systems are not always accurate, and many of us have probably asked questions akin to “why does Netflix keep recommending me Korean Dramas?”, or “why is Amazon trying to sell me a Kimono?”. The problem, originally highlighted by Herlocker et. al., involves stale or incorrect data, algorithm transparency and result explanation. If the mechanics and reasoning of a search or recommendation algorithm can be communicated to a user in the right way, it can improve acceptance of the prediction and trust in the system as a whole. Better still, by mapping the mechanics of information filtering algorithms into visual spaces, we support the notion of user control over the algorithm at recommendation time, helping to build trust, improve user experience and alleviate issues arising from stale or otherwise incorrect user profile information (such as those Korean Dramas and that Kimono…). This talk will explore current research at UCSB’s Four Eyes Lab that focuses on these questions. John O’Donovan will introduce the key concepts and challenges of interacting with artificial advice givers, and introduce a novel system known as MoodPlay, that recommends music based on an interactive latent space of mood information from a database of music artists. Byungkyu Kang will lead the second half of the talk, focusing on the ways that filtering and recommendation algorithms can be applied to social media data and represented in geo-spatial visualizations, and giving a discussion of the types of novel insights that these representations can provide.

odonovan-portraitJohn O’Donovan is an associate research scientist and principal investigator at the Computer Science Department, University of California, Santa Barbara. John received his PhD in Computer Science in 2008 from University College Dublin, Ireland, advised by Prof. Barry Smyth. His research background is in AI, with a focus on recommender systems. He has a particular interest in modeling trust in social networks, from the perspective of mining big network data, and also from the HCI perspective. Dr. O’Donovan has published more than 50 research papers in peer reviewed conferences and journals. His research on human computer interaction and intelligent interfaces has won multiple best paper awards, including at SocialCom 2013 and IEEE CogSima 2014 conferences. John has served on program committees and as reviewer for more than 20 conferences and journals, including ACM RecSys, ACM IUI, ACM CHI, WWW, KDD, IJHCS, ACM TOIT and ACM TIST. He is currently serving as general co-chair of the 2016 ACM international conference on Intelligent User Interfaces, and as an associate editor for ACM TiiS journal special issue on Human Interaction with Artificial Advice Givers.
kang-potraitByungkyu (Jay) Kang is a Ph.D. candidate in Computer Science at UCSB. He focuses on human computer interaction and is particularly interested in data mining, machine learning and visualization. His current research projects are (1) identifying reliable information in social media and (2) designing novel techniques for visual analytics. Recently, his work has explored potential reciprocal effects between quality of content and its visual representation in user information assessments in microblogs. He received his MS in Computer Science from UCSB and also has three years of experience in industry as a researcher and a system engineer, working at Samsung, IBM, KIST, Yahoo! Labs and Adobe Research.

The objectives of the ThinkSpatial brown-bag presentations are to exchange ideas about spatial perspectives in research and teaching, to broaden communication and cooperation across disciplines among faculty and graduate students, and to encourage the sharing of tools and concepts.
Please contact Andrea Ballatore (893-5267, to review and schedule possible discussion topics or presentations that share your disciplinary interest in spatial thinking.

Follow spatial@ucsb on Facebook | Twitter | Google+ | Google Calendar

ThinkSpatial: O’Donovan & Kang