Search Solutions 2013
27th November 2013
BCS London Office, UK
Dyaa Albakour,Udo Kruschwitz, Andrew MacFarlane,Tony Russell-Rose, John Tait and Nilesh Thatte(Eds),
BCS Information Retrieval Specialist Group
Programme
Page9:30 - 10:00 Registration and coffee
Session 1: The changing face of Search
10:00 / Introduction / 410:10 / Behshad Behzadi, Lead Technical Manager,Google Switzerland,Conversational Search / 6
10:35 / Wim Nijmeijer, Solution Consultant EMEA, Coveo, Complex enterprise search delivered in a day - no I'm not insane / 7
11.00 / Nikolaos Nanas, Co-Founder,NOOWITand Researcher atthe Institute of Research and Technology Thessaly in Greece,Adaptive Information Filtering: from theory to practice / 8
11.25 – 12.00 Coffee Break
Session 2: More than Search
12:00 / Henning Rode, Technical Lead Search,Textkernel,CV Search at TextKernel / 912:25 / Sandra Collins, Director,Digital Repository of Ireland,Information Preservation and Access at the Digital Repository of Ireland / 10
12:50 - 14:00 Lunch
Session 3: Understanding the User
14:00 / Filip Radlinski, Applied Researcher,Microsoft Research,Sensitive Online Search Evaluation / 1114:25 / Mounia Lalmas, Principal Research Scientist,Yahoo! Labs,An engaging click / 12
14:50 / Margaret Hanley, UX Consultant,Transform UK.Understanding users search intent / 13
15:15 - 15:45 Coffee Break
Session 4: Enterprise Search
15:45 / Helen Lippell, Strategic Metadata Manager,Press Association,Enterprise Search - How to triage problems quickly and prescribe the right medicine / 1416:10 / Peter Wallqvist, Commercial Director,Ravn Systems,Universal Search and Social Networking - Exploiting the features of each to enhance the other / 15
16.35 / Kristian Norling, Enterprise Search Evangelist,Findwise,Reflections on the 2013 Enterprise Search Survey / 16
Session 5: Fishbowl Session
17:00 / Hot Topics in Search
Want to discuss your own hot topic? Join in, it's a fishbowl session! / 17
Hall of Frame / 18
18:00 to 19:00 IRSG AGM
17:30 to 20.00 Drinks Reception
Introduction
Dyaa Albakour,Udo Kruschwitz,Andrew MacFarlane, Tony Russell-Rose,John Tait and Nilesh Thatte
BCS IRSG
Search Solutions is a special one-day event dedicated to the latest innovations in web & enterprise search. In contrast to other major industry events, Search Solutions aims to be highly interactive, with attendance strictly limited.
The Search Solutions story began in 2006, during the preparations for the ECIR conference which was held at Imperial College that year. Around that time there was a growing feeling among certain sections of the IRSG committee that the gap between the research and practitioner communities was growing ever wider. Moreover, it was apparent that although the IRSG had excellent links with the former, its engagement with the latter was relatively modest. The response was to propose an ‘Industry Day’ at ECIR 2006, which wouldbe a one-day satellite event aimed specifically at engaging the practitioner community andbuilding bridges between the two.
After some deliberation and planning, ECIR Industry Day finally took place on 13th April 2006, at BCS London, not far from the main event at Imperial College. It attracted a small but enthusiastic audience of some 50 people. The programme was modest in scale and ambition (just 8 speakers, no panel, and no social event) but the concept had clearly provenviable. Moreover, the idea of holding an annual eventaimed specifically at the practitioner community started to find favour with the wider IRSG membership.ECIR, however,was not the right vehicle for this, as its organisation and location moves around Europe on an annual basis. And so, the idea of Search Solutions was born.
This is the seventh year of the event, in which time we have enjoyed some 60+ talks from global players such as Google, Microsoft and Yahoo,alongside start-ups and SME’s such as Funnelback and Flax. The topics addressed have included all aspects of information retrieval, from more technical topics such as learning to rank to more user-focused topics such as query log analysis and user experience design. Coverage has also been extended to related issues such as knowledge organisation and digital libraries.
We have also grown the content of the event over this time. 2010, for example, saw the introduction of the panel/fishbowl session and the networking opportunities offered by the evening drinks reception. And in 2012 we added a tutorial programme,with fee waivers available this year for students who want to learn more from the foremost industry experts (see appendix 2). Despite these additions, Search Solutions 2013 remains true to its roots as a compact forum, with interactivity and innovation at its heart. In order to acknowledge the contribution over the years of our speaker to Search Solutions, we include a ‘Hall of Fame’ in appendix 1 at the end of the proceedings.
This year’sprogramme is divided into five sessions.
The first session focuses on the changing face of search. Behshad Behzadi provides a broader view by focusing on conversations in search. Nikolaos Nanas provides an overview of information filtering methods which breaks away from traditional IR methods.
The session following provides a couple of talks on wider issues in which search is used. Henning Rode looks at searching for CV, whilst Sandra Collins provides an overview of Digital Repositories and issues surround it such as preservation etc.
Session three focuses specifically on the users’. Filip Radlinski looks a using evaluation of search in real environments, Mounia Lalmas as user behaviour via Click data and Mags Hanley provides an overview of users intent.
The fourth session focuses on an issue which over the years has become something of the theme at Search Solutions, namely Enterprise Search. Helen Lippell looks at identifying problems in Enterprise Search whilst Peter Wallqvist looks at the relationship between Enterprise Search and Social Media (another hot topic in search these days). Kristian Norling provides an overview of a survey recently done on Enterprise Search, and some reflections on the implications of the results.
The last session is a fishbowl session, where attendees can voice their own views on issues which they regard as being the most important of the day – this may include any issues discussed during Search Solutions or could be something they think important enough to bring up in a public forum.
Behshad Behzadi,
Researcher,
Google Switzerland,
Conversational Search
In this talk, I will present how Search in Google is becoming more Conversational. I will demonstrate how in Conversational search users can talk naturally to Google as if they would talk to a person.
Wim Nijmeijer
Solution Consultant EMEA,
Coveo,
Complex enterprise search delivered in a day - no I'm not insane
Traditionally Enterprise Search systems have been synonymous with ERP systems, considerable amount of resource & investments and when the project is completed 3 years later - the service just doesn’t work. Coveo will show how complex enterprise search projects can be implemented and running within a day.
Nikolaos Nanas
Co-Founder, NOOWIT and Resesarcher
Institute of Research and Technology Thessaly in Greece
Adaptive Information Filtering: from theory to practice
Adaptive Information Filtering (AIF) seeks to provide an individual with content that is relevant to the individuals multiple and evolving interests. It is a complex and dynamic computational problem, with no established solution. After ten years of research and two years of development, Noowit is the first real world application of Nootropia, an immune-inspired user profiling model for AIF, that breaks away from traditional IR-based approaches. The talk will focus on the theoretical foundations of the model and the lessons learned from applying AIF in practice for the first time. The disrupting effect of content personalization on online publishing and media in general will also be discussed.
Short bio:
Nikolaos Nanas studied civil engineering at the Aristotle University of Thessaloniki (Greece). He received his M.S. degree in Intelligent Systems from the University of Sussex (U.K.) and his Ph.D. in Artificial Intelligence from the Knowledge Media Institute of the Open University (U.K.). His research work lies in the areas of Biologically Inspired Computing, Adaptive Information Filtering, Web Personalization, Complex Systems and Statistical Natural Language Processing. Since September 2011, he has been working on turning his academic work and experience into a real world solution to the Information Overload problem. Noowit is the result of this effort and the first personalized reading and publishing Web app, based on Content-based Document Filtering.
Henning Rode
Technical Lead Search
TextKernel
CV Search at TextKernal
CV Search deals with highly complex and structured queries. This comes as a challenge not only for the search engine, but also for the user interface. We show how we improved the search experience for our users, starting from a structured query language, making extensive use of facets and tag-clouds, towards auto-suggestion and automatic query generation. Since domain knowledge plays an important role for such interface features, we will also explain how we gather and enhance our knowledge resources.
Sandra Collins
Director,
Digital Repository of Ireland,
Information Preservation and Access at the Digital Repository of Ireland
DRI is charged with preservation and access to Ireland's cultural and social heritage. Knowledge discovery and integration is challenging in this domain. Humanities data can be rich and complex, non-standardised in format, without common or consistent metadata and ontologies, and can be subject to complex rights issues. Consensus and best practice regarding digitisation and metadata standards for common usage, that still retain the richness of different disciplines and data types, could enable open access to Humanities data, and facilitate data exchange and sharing between the wealth of archives, repositories and libraries across Europe.
Filip Radlinski
Applied Researcher
Microsoft Research,
Sensitive Online Search Evaluation
Online evaluation allows information retrieval systems to be assessed based on how real users respond to search results presented. Compared with traditional offline evaluation based on manual relevance assessments, online evaluation is particularly attractive in settings where reliable assessments are difficult or too expensive to obtain. However, the successful use of online evaluation requires the right metrics to be used, as real user behaviour is often difficult to interpret. In this talk I show how sensitive online evaluation can be conducted using the interleaving approach, comparing it to a variety of alternative online and offline approaches.
Mounia Lalmas
Principle Research Scientist
Yahoo! Labs
An engaging click
A good search engine is one when users come very regularly, type their queries, get their results, and leave quickly. With user engagement metrics from web analytics, these translate to a low dwell time, often low CTR, but a very high return rate. But user engagement is not just about this. User engagement is a complex phenomenon that requires a number of approaches for its measurement: we can ask the user about their experience though questionnaires, we can observe where they look or move the mouse, and we can calculate various web analytic metrics. For instance, questionnaires are hardly scalable but offer rich, qualitative insights, whereas click data can be collected on a large-scale but are more difficult to analyse. This talk will present various efforts aiming at combining approaches to measure engagement and seeking to provide insights into what makes an engaging experience in web search. The talk will focus of what makes users click or not click, and what this means in terms of user engagement.
This talk is based on a SIGIR 2013 industry talk given by Ricardo Baeza-Yates. This is joint work with Jahnette Lehmann, GeorgeDupret, Ilaria Bordino, Yelena Mejova, Ricardo Baeza-Yates and Elad Yom-Tov.
Margaret Hanley
UX Consultant,
Transform UK,
Understanding users search intent
For many years, understand users' intent in search has been focused on doing analysis of sentence structure and natural language processing. But this is hard to do when the majority of searches are 1-2 words.
In this talk, Mags Hanley will talk about how to understand users intent by:
- Reviewing search terms
- Understanding the content within the search domain
- Doing user testing
- Understanding how the organisation communicates to their users and how it impacts search
- Understanding how business priorities and the calendar year will change intent.
She will also discuss out how to improve search, once you understand the intent
Helen Lippell
Strategic Metadata Manager
Press Association,
Enterprise Search – How to triage problems quickly and prescribe the right medicine
The wail of "intranet search is rubbish!" is often heard in organisations large and small. With problems such as (but not limited to!) seemingly inflexible technology solutions, limited people resources and lack of clarity of business ownership, it can be daunting to think about making any improvements.This talk will present a methodology for triaging search problems and offering prescriptions for quick results that won't have nasty side-effects. "
Peter Wallqvist,
Commercial Director,
Ravn Systems,
Universal Search and Social Networking – Exploiting the features of each to enhance the other
Two trends are emerging with modern organisations: the use of universal/enterprise search systems to access enterprise content and the emergence of enterprise social networking platforms. These two technologies are most often deployed as two distinct systems with very little interaction between the two. Our experience have shown us that if you combine the two in a more tightly integrated service offering, many of the inherent drawbacks of each platform can be addressed using the features of the other system. This presentation will show how and provide tangible examples of how a search and social networking platform can work together to provide a superior user experience.
Kristian Norling,
Enterprise Search Evangelist,
Findwise,
Reflections on the 2013 Enterprise Search Survey
A year in the world of enterprise search might not be that long. And not much has changed since the survey 2012 results were presented, right?Kristian will reflect on the 2013 survey results from his layman perspective and compare Enterprise Search with… you’ll have to wait a few more weeks to hear with what.
Fishbowl session: "Hot topics in Search”
Want do discuss your own hot topic?
It’s a fishbowl session!
Appendix 1 – Speakers Hall of Frame
1
Alistair Duke / BT Labs / 2007Nick Patience / 451 Group / 2010
Ayse Goker / AmbieSense / 2008
Fernando Lucini / Autonomy / 2007
Conrad Taylor / BCS EPSG / 2007
Greg Lindahl / Blekko / 2010
Lewis Crawford / British Library / 2011
Ian Kegel / BT / 2011
David Phillips / Corpora Software / 2006
Marianne Sweeney / Daedalus Information Systems / 2011
Sandra Collins / Digital Repository of Ireland / 2013
Nicholas Kemp / DSTL / 2012
Tony Russell-Rose / Endeca / 2009
Dusan Rnic / Endeca / 2010
Alan Oliver / Ex Libris / 2009
Greg Grefenstette / Exalead / 2009
Tony Hart / Fast / 2006
Gora Sudindranath / Fast / 2007
Nick Townsend / Fast / 2008
Kristian Norling / Findwise / 2011
Charlie Hull / Flax / 2010
Matt Taylor / Funnelback / 2011
Richard Morgan / Funnelback / 2012
Till Kinstler / German Common Library Network / 2010
Leo Sauermann / Gnowsis / 2012
David Elworthy / Google / 2006
Thomas Hoffman / Google / 2007
Richard Russell / Google / 2009
Behshad Behzadi / Google / 2010
Martin Porter / Grapeshot / 2006
Lee Atkinson / iDNA Solutions / 2007
Mihai Lupu / Information Retrieval Facility / 2010
Martin White / Intranet Focus / 2012
Stella Dextre Clarke / ISKO UK / 2012
John Tait / jonhtait.net / 2011
Theo Huibers / KPMG / 2006
Elias Pampalk / Last.fm / 2008
Richard Boulton / Lemur Consulting
David Milward / Linguamatics
Wendy Johnson / MerseyBio
Steve Robertson / Microsoft
Nick Craswell / Microsoft
Milad Shokouhi / Microsoft
Mike Taylor / Microsoft
Vishwa Vinay / Microsoft
Gabriella Kazai / Microsoft
Filip Radlinski / Microsoft
Chirag Gandhi / mPhasis
Emma Bayne / National Archives
Nikolaos Nanas / NOOWIT
Clive Flashman / NPSA
Paul Matthews / Overseas Development Institute
Dave Mountain / Placr
Toby Mostyn / Polecat
Jarred McGinnis / Press Association
Helen Lippell / Press Association
Peter Wallqvist / Ravn Systems
Iain Fletcher / Search Technologies
Andrew Maisey / Solcara,
Unified Solutions
Roberto Cornacchia / Spinque
Thijs Westerveld / Teezir
Henning Rode / Textkernel
Margaret Hanley / Transform UK
Nigel Hamilton / Trexy
Rob Stacey / True Knowledge
Gary Jennings / WebOptimiser
David White / WebOptimiser
Nick Cox / Yahoo
Sian Clark / Yahoo
Vivian Lin Dufour / Yahoo
Ricardo Baeza-Yates / Yahoo
Alex Jaimes / Yahoo
Mounia Lalmas / Yahoo
1
Appendix 2 – List of Students awarded Fee waiver for Search Solutions 2013
Stuart James (University of Surrey)
Juan David Millan Cifuentes (City University London)
Silviu Paun (University of Essex)
Stuart Mackie (University of Glasgow)
Marc Sloan (University College London)
1