Third Workshop on Exploiting Semantic Annotations in Information Retrieval (ESAIR 2010)

Workshop Program October 30, 2010

The workshop will have a one-day program, with an introductory summary of results from previous workshops, a series of presentations, after lunch a break-out session, followed by a discussion on issues from the break-out sessions, facilitated by the organisers. The goal of the workshop is to produce a joint statement on future directions of purpose-driven semantic analysis, taking the challenge questions above as point of departure.

All accepted presentations will be given in the form of a poster and a brief booster presentation slot in the most relevant thematic session. Details of the program will be published here in the near future.

Keynote addresses

Accepted Presentations

Workshop Theme

The goal of this workshop is to create a forum for researchers interested in the use of semantic annotations for information access tasks. By semantic annotations we refer to linguistic annotations (such as named entities, semantic classes or roles, etc.) as well as user annotations (such as microformats, RDF, tags, etc.). The aim of this workshop is not semantic annotation itself, but rather the applications of semantic annotation to information access tasks on various levels of abstraction such as ad-hoc retrieval, classification, browsing, textual mining, summarization, question answering, etc.

Workshop history

• 1st ESAIR workshop, March 2008, in conjunction with ECIR 2008, Glasgow.
• 2nd ESAIR workshop, February 2009, in conjunction with WSDM 2009, Barcelona.

The previous two ESAIR workshops were exploratory workshops to discuss the research space around the topic - the most notable results from them will be published shortly in the journal Information Processing and Management (IPM), Volume 46, issue 4, July 2010. This third edition of the workshop intends to debate and propose future directions for the application of semantic annotations in practice.

Challenge questions

There is an increasing amount of structure on the Web as a result of modern Web languages, user tagging and annotation, and emerging robust NLP tools. These meaningful, semantic, annotations hold the promise to significantly enhance information access, by enhancing the depth of analysis of today's systems. Currently, we have only started exploring the possibilities and only begin to understand how these valuable semantic cues can be put to fruitful use.

Unleashing the potential of semantic annotations requires us to combine the insights of natural language processing (NLP) to go beyond bags of words, the insights of databases (DB) to use structure efficiently even when aggregating over millions of records, the insights of information retrieval (IR) in effective goal-directed search and evaluation, and the insights of knowledge management (KM) to get grips on the greater whole. The Workshop aims to bring together researchers from these different disciplines and work together on one of the greatest challenges in the years to come.

The desired result of the workshop will be concrete insight into the potential of semantic annotations, and in concrete steps to take this research forward; synchronize related research happening in NLP, DB, IR, and KM, in ways that combine the strengths of each discipline; and have a lively, interactive workshop where everyone contributes and which inspires attendees to think "outside the box."

To that end, we wish to see discussions at the workshop that will adress some of the following challenge points.

Application/Use Case

What use cases make obvious the need for semantic annotation of information? What tasks cannot be solved by document retrieval using the traditional bag-of-words? What are the prerequisites of successful application?

Annotation

What types of annotation are available? Are there crucial differences between author-, software-, user-, and machine-generated annotations? Named entities, temporal expressions on the one hand and sentiment and hedging on the other are examples of analyses beyond topic that have moved to profitable application. What is holding back the widespread use of these annotations? Are there other types of annotations that are within our grasp?

Result Aggregation

Whereas IR focuses almost exclusively on finding individual chunks of information, DB naturally focuses on results that combine information and produce aggregated results (think of OLAP queries), and KM naturally deals with the whole information space. How can we fruitfully combine these strengths?

Searcher/Query

With shallow 2.4 word navigational queries, there may be little benefit in semantic annotations. What expressive power is hidden in the semantic annotation? What is keeping searchers from exploring these powerful search request?

Organisers

Jaap Kamps, University of Amsterdam

Jussi Karlgren, SICS

Ralf Schenkel, MPI/Saarland University

Program committee

• Omar Alonso (Microsoft Bing, USA) (co-chair ESAIR 1 & 2)
• Pablo Castells (Universidad Autónoma de Madrid, Spain)
• Arjen de Vries (CWI, Netherlands)
• Shlomo Geva (Queensland University of Technology, Australia)
• Vanja Josifovski (Yahoo! Research, USA)
• Noriko Kando (National Institute of Informatics, Japan)
• Liz Liddy (Syracuse University, USA)
• Maarten Marx (University of Amsterdam, Netherlands)
• Paul Ogilvie (LinkedIn, USA)
• Hinrich Schütze (Universität Stuttgart, Germany)
• Andrew Trotman (University of Otago, New Zealand)
• Ozlem Uzuner (University at Albany, State University of New York, USA)
• Roman Yangarber (University of Helsinki, Finland)
• Hugo Zaragoza (Yahoo! Research, Spain) (co-chair ESAIR 1 & 2)