Controlling the complexity in comparing search user interfaces via user studies

Käki, Mika; Aula, Anne

doi:10.1016/j.ipm.2007.02.006

Cited by 23 publications

(19 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Your mother is interested in historical novels and travel (especially Central America). A within-group setup was used for two reasons (1) because this setup requires fewer participants and (2) because this setup is recommended by [14] because it tends to balance out individual differences between the participants' searching behavior.…”

Section: Discussion Of the Experimental Setupmentioning

confidence: 99%

Improving web search transparency by using a Venn diagram interface

Langer¹,

Frøkjær²

2008

Proceedings of the 5th Nordic Conference on Human-Computer Interaction: Building Bridges

View full text Add to dashboard Cite

The user interfaces of the most popular search engines are largely the same. Typically, users are presented with an ordered list of documents, which provide limited help if users are having trouble finding the information they need. This article presents an interface called the Venn Diagram Interface (VDI) that offers users improved search transparency. The VDI allows users to see how each term, or group of terms, in a query contributes to the entire result set of a search. Furthermore, it allows users to browse the result sets generated by each of these terms. In a test with 10 participants, the VDI was compared against a standard web search interface, Google. With the VDI, users were able to find more documents of higher relevance and were more inclined to continue searching. Their level of interactivity was higher, the quality of the answers they found was perceived to be better. Eight out of 10 users preferred the VDI.

show abstract

Section: Discussion Of the Experimental Setupmentioning

confidence: 99%

Improving web search transparency by using a Venn diagram interface

Langer¹,

Frøkjær²

2008

Proceedings of the 5th Nordic Conference on Human-Computer Interaction: Building Bridges

View full text Add to dashboard Cite

show abstract

“…Some studies have imposed time limits on each task to allow for a specific number of tasks within particular period of time (Over, 1997;Kaki and Aula, 2008). For instance, six tasks were included in the TREC-6 Interactive Track with a time limit of 20 minutes for each task (Over, 1997).…”

Section: Tasks and Topicsmentioning

confidence: 99%

An overview of semantic search evaluation initiatives

Elbedweihy

Wrigley

Clough

et al. 2015

Journal of Web Semantics

View full text Add to dashboard Cite

“…There are no set rules on how long one should give subjects to complete tasks; this is usually contingent on the type of task and study purpose. For instance, in an evaluation of Web search result surrogates, Käki and Aula [158]) imposed short time limits in an attempt to simulate how people actually scan surrogates in real life. In many other evaluations, subjects are given 10-15 min to complete search tasks.…”

Section: Timing and Fatiguementioning

confidence: 99%

“…The number of actions or steps taken to complete a task is another way to look at time and efficiency. Käki and Aula [158] formalize two time-based measures that have been used in IIR research, search speed and qualified search speed (Table 10.4). These measures are based on answers not relevant documents, but could be extended to cover this retrieval unit.…”

Section: Time-based Measuresmentioning

confidence: 99%