User Modeling in Heterogeneous Search Environments (HetUM 2015)
Workshop at CIKM 2015
The workshop is canceled

The workshop brings together current initiatives, resources and experts in user modeling and heterogeneous search to build a common ground for developing better user models in heterogeneous search environments. The workshop focuses on two questions: (1) how can we combine advances in user modeling and heterogeneous search; and (2) how can we build and share datasets that contain user search interaction data. The workshop program includes reports on related initiatives, presentations of state-of-the-art results in user modeling and heterogeneous search, and round-table discussions on the two main questions.

For more information on CIKM, please visit: http://www.cikm-2015.org.

Workshop at a Glance

When users interact with information retrieval (IR) systems, they leave rich implicit feedback in the form of clicks, etc. This feedback contains valuable information about users and about IR systems. Analyzing and interpreting user interactions and modeling user search behavior has become an important research direction. It enables us to better understand users, perform user simulations, improve search algorithms and build quality metrics.

However, IR systems become more and more heterogeneous: they deal with information of various media types, structure and semantics; run on multiple devices and support a variety of short- and long-term search tasks; serve users with different background and preferences. Better user models are needed to support effective IR methods in such heterogeneous environments.

Workshop Themes and Topics

The workshop has two main themes: (1) combining advances in user modeling and heterogeneous search in order to create better models of user search behavior in heterogeneous environments, and (2) building and sharing datasets that contain user search interaction data.

The first theme incorporates studies on click models, models of mouse movement, aggregated and federated search, etc. The second theme includes research on capturing and analyzing user search interactions, such as cursor trajectories and touch gestures. It also builds on experience in anonymizing query logs, organizing privacy reviews, etc.

The topics of the workshop include, but are not limited to:

  •   Modeling user behavior:

    • Click model construction

    • Modeling search interactions beyond clicks

    • Session-level user modeling

    • Cross-session and cross-device user modeling

    • Satisfaction prediction

  • Heterogeneous search:

    • Interaction with heterogeneous components on SERPs

    • Heterogeneous information source selection

    • Aggregated ranking of heterogeneous items

    • UI and UX design for heterogeneous search

    • Performance evaluation of heterogeneous search

  • Search interaction data mining:

    • Click-through data

    • Query reformulation data

    • Cursor movement data

    • Eye-tracking data

    • Other interaction data (e.g., purchase actions, location data, etc.)

  • Practical issues:

    • Acquiring sufficient amount of interaction data in academic environments

    • Acquiring interaction data at scale without disturbing the users

    • Conducting privacy reviews in academia and industry

    • Anonymizing search interaction data and releasing it to public

    • Reference/benchmark implementations for data acquisition, data analysis and user modeling

Workshop Organizers

Important Dates:

  • Submissions due: 19 June 26 June 2015 (UTC-12)
  • Special track for re-submissions of CIKM papers: 8 July 2015 (UTC-12)




Comments