guest ::
TY - THES AU - Franken, Sebastian TI - Supporting asynchronous, discontinuous, collaborative, complex search tasks by the visualization of search trails PB - RWTH Aachen VL - Dissertation CY - Aachen M1 - RWTH-2016-05020 SP - 1 Online-Ressource (XI, 316 Seiten) : Illustrationen, Diagramme PY - 2016 N1 - Veröffentlicht auf dem Publikationsserver der RWTH Aachen University N1 - Dissertation, RWTH Aachen, 2016 AB - To date, complex search tasks lack in support when it comes to asynchronous and discontinuous search in individual and collaborative search scenarios. Examples for complex search tasks are the planning of a family holiday trip or scientific research: These search tasks have in common that the overall result is composed of individual partial results, which strongly depend on the searchers’ personal preferences and want to be generated collaboratively. Especially when these web search tasks happen in an asynchronous or discontinuous way, current solutions provide only limited help. This thesis answers the research question ‘Can search trails provide support for complex web search and how should tool support look like?’. To achieve this, I develop a novel solution for supporting asynchronous, discontinuous, collaborative, complex web search tasks. I achieve this by implementing the web browser extension ‘SearchTrails’ which visualizes the user’s web search behavior as a search trail, visualized as a force-directed graph. The search trail resembles the course of the user’s web search activities, while the system 'SearchTrails' allows storing and exchange of the search process and its results. This way, saved search trails can be recreated by the users themselves, they can be exchanged between collaborating searchers and extended. My approach goes beyond existing approaches as it values the user’s individual search trail above generalized recommendations. Search trails as collaboration artifacts enable direct collaboration between users and provide an unfiltered insight where the collaborating searcher has searched before, and where results have been found. I show the effectivity and efficiency of my developed approach in two user studies. The first user study qualitatively shows the effectiveness of the developed approach, while the second user study quantitatively shows its efficiency. The second user study especially focuses on collaborative search scenarios and shows the impact of the concept of search trails on the quality of the collaborative search process and its search results. Based on the findings from the user studies, the main research question can be answered positively. LB - PUB:(DE-HGF)11 UR - https://publications.rwth-aachen.de/record/659882 ER -