BCS Office, 1st Floor The Davidson Building 5 Southampton Street London, WC2E 7HA [Location]
Search Solutions is a joint event of ISKO UK and BCS IRSG (Information Retrieval Specialist Group). It is the premier UK forum for presentation of the latest innovations in search and information retrieval. In contrast to other major industry events, Search Solutions aims to be highly interactive, with attendance strictly limited.
The programme includes presentations, panels and keynote talks by influential industry leaders on novel and emerging applications in search and information retrieval.
The morning of the second day of the Search Solutions Conference (26th November) was an interesting mixture of demonstrations of the not-so-distant future (voice-based search from Google), technical descriptions of text embedding in sponsored search (Yahoo! and Microsoft), and in contrast a look at user needs and how to test whether they are being met. The user perspective was provided by Tony Russell Rose, who surprised many of us by reporting that people who work in recruitment are the ‘Boolean black belts’ of search, needing to master complex queries to pinpoint the skills and experience they’re looking for. The following talk from Charlie Hull (Flax) covered the general misconception that ‘search is magic’, and described some methods for testing the effectiveness of particular approaches.
Alison Weightman from Cochrane spoke about the use of text mining in literature reviews. Tom Crane (Digirati) and Dave Clarke (Synaptica) talked about different aspects of image retrieval, Tom focussing more on the International Image Interoperability Framework (which provides APIs for exposing images and metadata), while Dave built on the great presentation he gave at our last Conference, describing the use of Linked Data and semantic annotations to make images and parts of images retrievable from all kinds of perspectives.
Finally we also heard from Tessa Radwan of the Newspaper Licensing Agency, who spoke about the retrieval of news items in all their published versions (i.e. not just as intermediated text, but also as the original publication), while David Corney of Signal Media talked about approaches for identifying duplicate news items within their mammoth collection of c250 million articles.
There was the usual good turnout for the day (c75 people), and the excellent catering contributed greatly to the networking and general enthusiastic atmosphere.