Interdisciplinary Approaches to User-driven Exploratory Search and Interactive Data Analytics


Organizations from across academia, government and the commercial sector have realized for some time, the intrinsic value in the provision of user-driven search and data analysis. Information retrieval techniques operating on text or semantic annotations, have become the industry standard for retrieval from large data collections, such as documents, images, videos, music, medical data, etc. Implementations associated with such techniques work well with sufficient volumes of high-quality meta-data or tagging however it has become obvious that the use of subject matter expert (SME) tagging services simply does not scale with today’s (and future) projected data volumes. This is particularly evident within the Earth Sciences domain where SME’s are constantly in high demand and oversubscribed. This combined with the sheer volume of data belonging to ever growing datasets means that our community must move towards more adaptive intelligent mechanisms for ensuring current search and data analysis requirements are addressed.
A growing recognition exists that a solution to this problem involves active engagement of the user (on several levels) in the overarching, multifaceted search and analysis workflows which comprise of information retrieval, exploratory search and interactive data analytics. Thus the goal is to enable users to not only explore given dataset(s) more easily but also gradually direct their search to a more specific area of the search space.
The aim of this session is to explore new methods and interface/system design for interactive data analytics, user-driven exploratory search, search customization and related management in various domains, including specialized scientific collections e.g. datasets, virtual collections, scientific literature as well as for various tasks, such as semantic information retrieval, conceptual organization and clustering of data collections for sense making, semantic expert profiling, and document/multimedia recommender systems. The primary audience of the session are researchers and practitioners in the area of interactive and
personalized system design possibly relating to data management and systems technologies, data archival and stewardship and/or new search paradigms as applied to the Earth Sciences who are interested in discussing these topics in a multidisciplinary setting.
Session Target Areas
The workshop invites submissions in all areas of interactive data analytics and exploratory search including:
- design, testing and assessment of interactive systems for data analytics
- interactive data visualization for exploratory and investigative analysis
- interactive classification and clustering of data
- user assisted curation and validation of the analysis process, and generation of data visualizations
- user engagement in the semantic analysis process via suitable annotation and correction tools
- study of the trade-off between accuracy of the results and user effort
- personalization and user modeling related to interactive system design
- user tagging
- provision of user-specific content
- interactive user feedback
- mining logs (HTTP, FTP, etc.)
- recommendation systems
- expert based rule systems 

Submissions aimed at discussing or solving practical problems from within the Earth Sciences are especially welcome.
We encourage submissions of work in progress, concepts, case studies, and generally material that will stimulate discussion, generate useful feedback to the authors, encourage research collaborations and vigorous exchange of ideas on promising research directions.

Mcgibbney, L.; Neufeld, D.; Armstrong, E.; Interdisciplinary Approaches to User-driven Exploratory Search and Interactive Data Analytics; Winter Meeting 2017. ESIP Commons , October 2016