Program

Tuesday November 11

Time	Event
08:30 - 09:30	Registration
09:30 - 09:45	Welcome
09:45 - 11:15	Session 1: Domains (chair: Paul Buitelaar) Combined Language Processing Methods and Mash-Up System for Improving Retrieval in Diabetes Related Patents (Ivan Chorbev, Danco Davcev and Daniela Boshnakoska) News Articles Classification Using Random Forests and Weighted Multimodal Features (Dimitris Liparas, Yaakov HaCohen-Kerner, Anastasia Moumtzidou, Stefanos Vrochidis and Ioannis Kompatsiaris) When Text Authors Lived Using Undated Citations (Dror Mughaz, Yaakov HaCohen-Kerner and Dov Gabbay) Temporal Context for Authorship Attribution (Niels Dalum Hansen, Christina Lioma, Birger Larsen, and Stephen Alstrup)
11:15 - 11:45	Coffee
11:45 - 13:00	Panel: Professional Search in a Modern World (chair: Kalervo Järvelin)
	Panel members: Fabio Crestani, Gabriella Pasi, Christian Boitet, Stephane Marchand-Maillet, Elaine Toms
13:00 - 14:00	Lunch
14:00 - 15:00	Keynote: "Don’t Hurt Them: Learning to Rank from Historical Interaction Data" by Maarten de Rijke) (chair: Paul Buitelaar)
15:00 - 15:30	Coffee
15:30 - 17:00	Session 2: Systems and algorithms (chair: Stephane Marchand-Maillet) Fast Interactive Information Retrieval with Sampling-Based MDS on GPU Architectures (Hasmik Osipyan, April Morton and Stéphane Marchand-Maillet) Which One to Choose: Random Walks or Spreading Activation? (Serwah Sabetghadam, Mihai Lupu and Andreas Rauber) Post OCR correction of Swedish Patent Text (Linda Andersson, Helena Rastas and Andreas Rauber) Extracting Nanopublications from IR Papers (Aldo Lipani, Florina Piroi, Linda Andersson and Allan Hanbury)
18:30 - 22:00	Conference dinner at Restaurant SULT

Wednesday November 12

Time	Event
08:30 - 09:00	Registration
09:00 - 10:30	MUMIA Management Committee Meeting & Evaluation –- Reporting 1 (chair: Michail Salampasis)
	Action Chair and Grant Holder presentations Presentations from Working Group leaders
10:30 - 11:00	Coffee
11:00 - 12:20	MUMIA Management Committee Meeting & Evaluation –- Reporting 2 (chair: Andreas Rauber)
	STSMs, IR/NLP/MLT integration, Standards & Protocols, Dissemination, Training Schools, Industry and Organisation Impact, Early Stage researchers, and Case Studies.
12:20 - 13:00	Discussion and Questions
13:00 - 14:00	Lunch
14:00 - 15:30	Session 3: Design and evaluation (chair: Allan Hanbury) IPC Selection Using Collection Selection Algorithms (Anastasia Giachanou and Michail Salampasis) Using Eye-Tracking to Investigate Patent Examiners’ Information Seeking Process (Fernando Loizides and Barrou Diallo) An Evaluation of an Interactive Federated Patent Search System (Michail Salampasis, Anastasia Giachanou and Allan Hanbury)
15:30 - 16:00	Coffee
16:00 - 17:00	Panel: The Next Steps after the MUMIA COST Action (chair: John Tait)
	Panel members: Fernando Loizides, Mihai Lupu, Laurentiu Vasiliu, Mike Salampasis

You can find the PDF version of the agenda of the MUMIA Management Committee Meeting & Evaluation here.

Keynote

Don’t Hurt Them: Learning to Rank from Historical Interaction Data
Maarten de Rijke (University of Amsterdam)

One of the main advantages of online evaluation schemes is that they are user-based and, as a result, often assumed to give us more realistic insights into the real system quality than off-line methods. This is also one of their main disadvantages: comparing two rankers online requires presenting users with result lists based on those rankers and observing how users interact with them. New rankers may perform sub-optimal and hence hurt the user experience. Can we use or re-use historical data, collected from user interactions with a production system, to assess or optimize new alternative rankers? This question has increasingly gained interest in the past few years. In the talk I will contrast several proposals for learning from historical interaction, based on importance sampling, random buckets, and a Bayesian approach based on explicit user models.

This is based on joint work with Artem Grotov, Katja Hofman, Damien Lefortier, Anne Schuth, Shimon Whiteson.

Maarten de Rijke is full professor of Information Processing and Internet in the Informatics Institute at the University of Amsterdam. He holds MSc degrees in Philosophy and Mathematics (both cum laude), and a PhD in Theoretical Computer Science. He worked as a postdoc at CWI, before becoming a Warwick Research Fellow at the University of Warwick, UK. He joined the University of Amsterdam in 1998, and was appointed full professor in 2004.

De Rijke leads the Information and Language Processing Systems group, one of the world's leading academic research groups in information retrieval. During the most recent computer science research assessment exercise, the group achieved maximal scores on all dimensions. His research focus is on intelligent information access, with projects on self-learning search engines, semantic search, and social media analytics.