Keynote Talks

Sue Dumais
Susan Dumais (Microsoft Research, Redmond)
Bio: Susan Dumais a Distinguished Scientist at Microsoft and Deputy Managing Director of the Microsoft Research Lab in Redmond. Prior to joining Microsoft Research, she was at Bell Labs and Bellcore, where she worked on Latent Semantic Analysis, techniques for combining search and navigation, and organizational impacts of new technology. Her current research focuses on user modeling and personalization, context and search and temporal dynamics of information. She has worked closely with several Microsoft groups (Bing, Windows Desktop Search, SharePoint, and Office Online Help) on search-related innovations. Susan has published widely in the fields of information science, human-computer interaction and cognitive science, and holds several patents on novel retrieval algorithms and interfaces. Susan is also an adjunct professor in the Information School at the University of Washington. She is Past-Chair of ACM's Special Interest Group in Information Retrieval (SIGIR), and serves on several editorial boards, technical program committees, and government panels. She was elected to the CHI Academy in 2005, an ACM Fellow in 2006, received the SIGIR Gerard Salton Award for Lifetime Achievement in 2009, was elected to the National Academy of Engineering (NAE) in 2011, and received the ACM Athena Lecturer Award in 2014.
Abstract: Understanding a short search query in isolation is a very difficult task. Query understanding is much easier if we consider the "context" in which the query arises, e.g., previous queries the searcher has issued, the current location and time, etc. Traditionally search engines have returned the same results to everyone who asks the same question. However, using a single ranking for everyone, in every context limits how well a search engine can do. I begin by outlining a framework to characterize the extent to which different people have the same (or different) intents for a query. I then describe several examples of how we represent and use context to improve search quality. Finally I conclude by highlighting some challenges in developing contextually-aware algorithms including evaluation.

Fabio Ciravegna
Fabio Ciravegna (University of Sheffield)
Bio: Fabio Ciravegna is Professor of Language and Knowledge Technologies at the Department of Computer Science at the University of Sheffield. His research field concerns Knowledge and Information Management over large scale, covering 3 main areas: (i) How to capture information over large scale from multiple sources and devices (the Web, the Social Web, distributed organisational archives, mobile devices, etc.), (ii) how to use the captured information (e.g. for knowledge management, business intelligence, customer analysis, management of large scale events, etc.); and (iii) how to communicate the information (to final users, problem owners, etc.). He is the director of the European Project WeSenseIt on citizen observatories of water and principal investigator in the EPSRC project LODIE (Large Scale Information Extraction using Linked Open Data). He has developed with Neil Ireson and Vita Lanfranchi methodologies for event monitoring in social media that have been used to support the emergency services and organisers in several large scale events involving hundreds of thousands of people; among them the Glastonbury Festival (200,000 participants), the Bristol Harbour Festival (250,000), the Tour de France (UK part), the evacuation of 30,000 people from the City of Vicenza and many others. He has co-created two companies: K-Now Ltd who commercialises the social media analysis technology and The Floow Ltd who develops technology currently monitoring hundreds of thousands of drivers for motor insurance via mobile phones.
Abstract: The ubiquitous use of mobile devices and their use for social activities make possible to see events and their development through the eyes and the senses of the participants. In this talk I will discuss my experience in working with emergency services and organisers of very large events involving hundreds of thousands of participants to help identify planned and unplanned situations through social media. This involves analysis of social media messages (Twitter, Facebook, etc.) as part of the tasks of the emergency service control room. Applications range from tackling natural and man-made disasters (floods, earthquakes, large fires, etc.), to overseeing very large events such as City and Music Festivals. The task requires high focus on the geographic area, understanding of the social context and the event nature, as well as instinct and experience to cope with large crowds and their sometimes erratic behaviour. It is fundamentally a human centred task that requires important support by computers, as long uncomfortable shifts may be involved (sometimes 24/7) and the amount of material to cope with can be huge (millions of messages and pictures to shift through). From a technical point of view, this support requires real-time large-scale text and data analysis, visual analytics and human computer interaction. In this talk I will discuss the requirements for this support, focussing mostly on the social media analysis part. I will discuss the issues, some of the current technical solutions and a roadmap for future development.

Ann Blandford
Ann Blandford (UCL, London)
Bio: Ann Blandford is Professor of Human-Computer Interaction at UCL and former Director of UCL Interaction Centre. Her first degree is in Maths (from Cambridge) and her PhD in Artificial Intelligence (from the Open University). She started her career in industry, as a software engineer, but soon moved into academia, where she developed a focus on the use and usability of computer systems. She leads research on how people interact with and make sense of information, and how technology can better support people's information needs, with a focus on situated interactions. She has over 200 international, peer-reviewed publications, including a Synthesis Lecture on "Interacting with Information".
Abstract: Most evaluations of information access systems put the system at the centre. In this talk, I will present an alternative focus: on the person, who may be actively seeking, or may simply be encountering information as they engage with a variety of information resources. The "information journey" is a way of framing our understanding of both directed seeking and encountering. This framework was developed based on empirical studies of people interacting with information across a variety of contexts. This perspective emphasises the importance of evaluating the user experience of interacting with systems, and how they support the individual's information journey. The system has to support people's work, as well as being effective and usable. In this talk, I will present the information journey, relating it to models of information retrieval and information seeking. I will then present approaches to evaluating information access systems that start from the premise that every evaluation study has a purpose, that the study design has to be appropriate for that purpose, and that, ultimately, the success of any system depends on how it is used by people engaged in information work in their daily lives.