The dynamic HomeFinder

Abstract

We designed, implemented, and evaluated a new concept for visualizing and searching databases utilizing direct manipulation called dynarruc queries. Dynamic queries allow users to formulate queries by adjusting graphical widgets, such as sliders, and see the results immediately.By providing a graphical visualization of the database and search results, users can find trends and exceptions easily. User testing was done with eighteen undergraduate students who performed significantly faster using a dynamic queries interface compared to both a natural language system and paper printouts. The interfaces were used to explore a real-estate database and find homes meeting specific search criteria.

Conceptual and Methodological Issues in Evaluating Multidimensional Visualizations for Decision Support

“…In one scenario, a user browses cameras to buy, by creating a lasso query and refining it while transitioning between scatterplots [30]. HomeFinder represents data cases as dots on a map, while other attributes are represented as dynamic query widgets that can be used to progressively refine a query [84]. FilmFinder generalizes HomeFinder by changing the map into a scatterplot display [2].…”

“…Some of their design features have been assessed through qualitative studies, but without comparison to other techniques [6][7][8]37,61,64,89]. The few exceptions are controlled experiments comparing either variations of the same visualization [11,24], or comparing a visualization with non-visualization base cases, such as web forms [67], static numerical tables [85], or Q&A systems and textual formats [84].…”

“…Some studies have examined subjective user ratings [8,61,67]. More formal evaluations have employed tasks such as value retrieval [11,24,84], range tasks [67,84], finding extrema [11,24,89], finding outliers [84], and identification of patterns [84], correlations [89], and clusters [89]. Other studies involved more complex analytic tasks combining multiple low-level tasks [8,67].…”

“…However, to our knowledge, no previous work has evaluated such tools for their ability to support decision-making tasks. Most existing studies are either qualitative studies without a comparison baseline [6-8, 37, 61, 64, 89] as value retreival [55,84] or correlation estimation [57,89] instead of decision-making tasks. Although supporting elementary analytic tasks is likely an important precondition for supporting informed decisions, data-driven decision making differs from data analysis and data exploration.…”

What is Interaction for Data Visualization?

“…This action, as well as painting on, or adding stickers to Zooids, illustrate that high flexibility in interaction means allows for appropriation of a data interface. Category -Data actions: we compare the HomeFinder [122] and the constructive visualization approach [59] colored tiles ("Token"). The first example focused on cardinality of interaction means.…”