TY - GEN
T1 - Chapter 13
T2 - 1st Workshop on Search Computing Challenges and Directions, SeCo 2009
AU - Bozzon, Alessandro
AU - Brambilla, Marco
AU - Ceri, Stefano
AU - Fraternali, Piero
AU - Manolescu, Ioana
PY - 2010
Y1 - 2010
N2 - Liquid queries are a flexible tool for information seeking, based on the progressive exploration of the search space; they produce "fluid" results which dynamically adapt to the shape of the query, as a liquid adapts to its container. The liquid query paradigm relies on the SeCo service mart and multi-domain query execution concepts: an expert user selects a priori the service marts relevant to the information seeking task at hand and the connections necessary to join them, and publishes such a definition in the SeCo back-end. The Liquid Query client-side interface consumes the application definition created by the expert and dynamically builds a query interface for the end-user. Such interface allows one to supply keywords to query the pre-configured service marts and offers controls for exploring the combinations computed by the SeCo execution engine. The interaction commands are based on a tabular representation of results and comprise: reordering, clustering, addition or deletion of attributes, addition of extra service marts to the query for specific items in the result set or for the entire result set, request of more results from all services or from selected ones, expansion of details on selected items, and more. The Liquid Query is equipped with multiple data visualization options suited to render multi-domain results and can be instrumented with indicators showing the quality of the result set.
AB - Liquid queries are a flexible tool for information seeking, based on the progressive exploration of the search space; they produce "fluid" results which dynamically adapt to the shape of the query, as a liquid adapts to its container. The liquid query paradigm relies on the SeCo service mart and multi-domain query execution concepts: an expert user selects a priori the service marts relevant to the information seeking task at hand and the connections necessary to join them, and publishes such a definition in the SeCo back-end. The Liquid Query client-side interface consumes the application definition created by the expert and dynamically builds a query interface for the end-user. Such interface allows one to supply keywords to query the pre-configured service marts and offers controls for exploring the combinations computed by the SeCo execution engine. The interaction commands are based on a tabular representation of results and comprise: reordering, clustering, addition or deletion of attributes, addition of extra service marts to the query for specific items in the result set or for the entire result set, request of more results from all services or from selected ones, expansion of details on selected items, and more. The Liquid Query is equipped with multiple data visualization options suited to render multi-domain results and can be instrumented with indicators showing the quality of the result set.
KW - Exploratory search
KW - Search computing
KW - User interfaces
UR - http://www.scopus.com/inward/record.url?scp=77951585122&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-12310-8_13
DO - 10.1007/978-3-642-12310-8_13
M3 - Conference contribution
AN - SCOPUS:77951585122
SN - 3642123090
SN - 9783642123092
VL - 5950 LNCS
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 244
EP - 267
BT - Search Computing - Challenges and Directions
PB - Springer
Y2 - 17 June 2009 through 19 June 2009
ER -