An investigation of the query behavior in task‐based collaborative exploratory Web search

Abstract

Collaboration in the information seeking and retrieval environment is common, particularly when the search task is complex and exploratory. Multiple factors such as contextual features and task type can affect users' query behavior. This paper presents a study investigating the effects of collaboration and task types on users' query behavior. The study involves two conditions: collaborative search and individual search, and the two search tasks: the recall‐oriented information‐gathering and the utility‐based decision‐making. We analyze users' query behavior in three dimensions: basic query features (e.g. the number of queries), query reformulation patterns (e.g. New, Specification, Generalization and Reconstruction) and query performance. The findings of this study reveal that queries are more diverse in collaborative search and recall‐oriented tasks. Users employed New and Specialization more often as query reformulation types in collaborative search while people in individual search use Reconstruction more often. Besides, the successful query rate is higher in individual search and recall‐oriented tasks.

An investigation of the effects of awareness and task orientation on collaborative search

“…The lack of differences we observed in the traditional information retrieval metrics of precision and recall suggest that studies of collaborative search should consider more than these outcome measures. Collaborative search metrics are still being defined and developed – e.g., (Shah & González‐Ibáñez, ; Yue et al, ) – and additional measures are needed. The behavioral concordance measures we developed were effective in finding interesting effects of awareness, and provide another set of tools for use in understanding collaborative search behaviors.…”

“…Collaborative search metrics are still being defined and developed -e.g., (Shah & González-Ibáñez, 2011;Yue et al, 2013) -and additional measures are needed. The behavioral concordance measures we developed were effective in finding interesting effects of awareness, and provide another set of tools for use in understanding collaborative search behaviors.…”

Factors that influence query reformulations and search performance in health information retrieval: A multilevel modeling approach

“…Kinley, Tjondronegoro, Partridge, and Edwards (2012) found that users needed to reformulate more queries in exploratory tasks requiring high search skills. Yue, Han, and He (2013) found that system types would influence users' query reformulation behavior. Xie and Joo (2012) found that users were more likely to modify their queries when using OPACs than when using search engines.…”

“…Yue et al (2013) found that users added more new terms in an academic literature search than in a travel information search. For example, found that specialization was observed frequently in simple and hierarchical tasks, whereas substitution was applied more frequently in parallel search tasks.…”

“…Xie and Joo () found that users were more likely to modify their queries when using OPACs than when using search engines. Yue, Han, and He () found that system types would influence users' query reformulation behavior. They found that users were able to come up with better query reformulation strategies in a new interface by using search histories.…”