Adaptive Search Support for Information Seeking Stages
Hugo Huurdeman · UvA-DARE (University of Amsterdam) · 2015
We use the Web for work, leisure, and research, assisted by various search systems in the task of satisfying our information needs.We utilize these systems to perform our daily tasks, ranging from simple lookup tasks to complex, exploratory and analytical ventures.The more complex tasks may involve multiple information seeking stages, with evolving inherent needs for each stage.Most search systems, however, only support these complex tasks in an elementary manner, and offer a 'one-size-fits-all' interface optimized for shallow lookup search.In addition to the wealth of information available on the live Web, historical Web content is currently available in Web archives, containing snapshots of the Web that once was.These Web archives can enable new opportunities for analytical tasks, serving as data sources for researchers in various fields.At the moment, however, few archives offer full-text search, and the search systems that are available fall short of the rich functionality needed for analytical tasks.This PhD research proposal takes Kuhlthau's ISP model as its framework, and addresses search support for different 'stages' of complex search tasks.It discusses the theoretical implications of multistage information seeking models for the design of search systems.The proposal examines the effects of information seeking stages on the flow of interaction with actual search systems.The understanding on both the theoretical and practical level are used to design and evaluate multistage search systems, firstly in a general Web search setting, and secondly in a Web archive search setting.Finally, this leads to design recommendations for supporting different stages of complex tasks in search systems.This proposal consists of five parts: first, we introduce the topic and research problem.Then, we state the research questions.Subsequently, background literature is briefly discussed, followed by the employed methodology.Next, current progress is discussed, before summarizing this proposal in the conclusion.