The Search Engine Meeting

April 26-27, 2010Hyatt Regency Boston • Boston, MA
(Preconference Workshops: Sunday, April 25)
Program [PDF]
Daily Schedule
Day One
Day Two
Attendee Survey
Past Shows




General Conference - Day One: Monday, April 26, 2010
PreConference Day One Day Two
Continental Breakfast
8:00 am – 9:00 am
Old Search, New Search: Adapting to New Realities
9:00 am – 10:00 am
Bipin Patel, ProQuest

Search was not invented as recently as internet denizens suppose. Computer-based information retrieval not only predates the internet; it predates the personal computer. Search systems such as Dialog started more than 30 years ago. The challenge is to modernize search technologies in legacy products to make them appealing and useful to today’s searching community, which is vastly larger and more diverse than it was decades ago. Bipin Patel, CIO of ProQuest and formerly CIO of Ford Motor Co., confronts not one, but three legacy search platforms—Cambridge Scientific Abstracts (CSA), Dialog, and ProQuest. Using personas, Patel and his group are recreating search for the 21st century. New platforms must take user expectations into account, including those about search simplicity and relevant results.

10:00 am – 10:30 am
Implementing Search on a Massive Scale
10:30 am – 11:00 am
Christian Vogt, Raytion

What to do with tens of thousands of users, billions of documents, and petabytes of data? How can flexible search platforms be built that scale across the whole organization and that offer search as a global resilient service? Retrieving information for thousands of users from enterprise systems distributed around the globe and containing a huge number of documents is not a trivial task. Despite varying user group requirements and the complexity of technical infrastructures, there are basic principles that make such search solutions possible. This session presents the essential aspects that have to be taken care of when implementing very large-scale-search applications and provides basic principles for building such platforms.

THESEUS and Its Impact on the Semantic Search Marketplace
11:00 am – 11:30 am
Martin Baumgärtel, Client Services, Velocitude

The German government started an R&D initiative in 2006 to develop search and semantic technologies to empower domestic companies to play an important role globally. An independent analysis will shed light on research objects, results, the project’s track, and determine whether the original goals are being met. The focus is on core players in the project and the potential impact on the semantic search marketplace. The project should be measured against the initial goals and the budget size.

Semantic Search Engines and Biomedical Knowledgebases for Consumer Health
11:45 am – 12:15 pm
Tamas Doszkocs Ph.D., President, WebLib

There are many specialized traditional health search engines, such as and, that provide high-quality consumer health information. Unfortunately, most of these sites are limited in their scope of coverage and search functionality. The current generation of semantic health search engines attempts to cover the full spectrum of health concerns, however, these engines vary widely in the quality of their knowledgebases and the sophistication of their semantic search capabilities. This presentation will survey the state of the art and will demonstrate the cutting-edge research and operational implementations at the National Institutes of Health and in the commercial sector.

Mobile Search
12:15 pm – 12:45 pm
Dr. David A Evans, Evans, LLC

As mobile devices proliferate, with some pundits proclaiming that the devices will be the computer of the future, replacing our current desktop and laptop machines, designing search for such tiny screens has become increasingly important. What elements of the user interface can be optimized for mobile search and how can useful results be presented? Are there principles that can be applied to UI design?

12:45 pm – 2:00 pm
Innovations in Social Search
2:00 pm – 2:30 pm
Jeff Fried, BA Insight, USA

Search can both fuel and tap social behavior. How can social networks make the core elements of search more relevant? This talk will cover innovations in three areas: techniques for finding people and expertise; approaches for using social behavior to improve relevancy ranking; and collaborative-searching experiences. What are the special techniques involved in co-searching? Fried will demonstrate experimental systems in each of these areas, as well as covering algorithmic approaches to social search. He will also show how this technology applies in business facing applications such as SharePoint.

The Wow in Now: Social Search’s Gravitational Pull in Information Retrieval
2:30 pm – 3:00 pm
Stephen E. Arnold,

Social search—the indexing of content flowing from,, RSS feeds, and other real-time content—has emerged as a “must have.” Enterprise and web search systems have struggled to provide timely, relevant results to queries across these content geysers. The benefit of real-time search rests upon the belief that content processing systems can add context to often brief, sometimes inaccurate, and frequently densely backlinked content objects. Commercial services rely on numerical algorithms and brute force string matching to deliver useful items germane to a query. The downside is that the information filtered from the content gushers may lack context, have unknown provenance, and can be excessively self-referential. The talk concludes with observations about relevance ranking across the tested system.

3:00 pm – 3:30 pm
The Semantic Web Landscape
3:30 pm – 4:00 pm
Leo Obrst, The MITRE Corp

This session, presented by Leo Obrst, MITRE’s principal artificial intelligence scientist, information semantics, will provide an overview of the current landscape of semantic web products and services and highlight a few recently released products that utilize Semantic Web technologies.

Universal, Composable Indexing
4:15 pm – 5:00 pm
Chris Biow, MarkLogic Corporation

This presentation demonstrates a radical extension of special-purpose XML databases, incorporating conventional text indexing, “universal” indexing of structure and XML semantics, scalar range indexes, spatial indexes, and “reverse” indexes of search expressions. These “forward” and “reverse” indexes reside within the same  atabases, in the same search engine and DBMS kernel, and use the same extended XQuery language. Indexed databases are built on XML documents that may contain arbitrary combinations of structured or semistructured data, textual content, geospatial locations, and search expressions. Queries may then be run against such databases, where the queries themselves contain arbitrary XML documents mixing text and data, full-text and semantic search expressions, and geospatial constraints. This creates a true “data finding data” capability, in which search expressions are simultaneously evaluated in both directions between the database and the query under evaluation.

5:00 pm – 6:00 pm
Lanyard/Badge Sponsor
Bag Sponsor
Media Sponsors


© 2009 - 2017, Information Today, Inc. Privacy/Cookies Policy