Visual Information Retrieval Interfaces

Bob Korfhage (deceased) and Michael Lewis

Department of Information Science University of Pittsburgh 135 N. Bellefield Ave Pittsburgh, PA 15208

CONTACT INFORMATION

ml@lis.pitt.edu (412) 624-9426 fax: (412) 624-2788

PROGRAM AREA

Usability and User-Centered Design

KEYWORDS

visualization information retrieval usability visual inference

PROJECT SUMMARY

Despite decades of exposure to Boolean queries many of the public remain unsure of their semantics. Vector based retrieval, as presented on the Web, is even more opaque, allowing users to search without knowing how the terms they type in are combined or used. Visual information retrieval interfaces (VIRIs) have been developed by researchers as a more natural representation of multidimensional queries. VIRIs had been believed to make the tasks of query formulation and interpretation easier particularly for inexperienced users. When a prototypical VIRI, VIBE, was tested using professional and naive searchers, however, it was neither preferred nor did it improve performance (Koshman 1996).

We hypothesized that the sophistication and proliferation of features in the experimental VIRIs imposed a burden of learning and cognitive load which obscured the simplicity and usefulness of the visual inferences these displays supported. In the first year of the grant we demonstrated that drastically simplified versions of two such VIRIs (WebVibe and J-BIRD) could in fact enhance performance for naive subjects.

In a series of subsequent studies, we have been examining subjects' performance at information retrieval inference tasks using canonical visual representations. These included ordered word lists, tables, multidimensional icons, scatterplot graphs, and proportional attraction using VIBE's "spring" analogy. These studies have ranged from in-class paper and pencil tests, to monitored performance in computer labs, to web-based data collection in both the US and Norway. Results for 2 and 3 term Boolean and vector-based queries suggest that users prefer visual representations of queries and results, do no worse with visual presentations, and perform better using visual presentations for higher (3 "term") dimensional queries.

In a parallel field evaluation WebVibe was used as an alternate retrieval interface to a web accessible database of roughly 4,000 business contracts over a two month trial in 1998. Comments and anecdotal support experience and usage records suggest users found the VIRI helpful in accessing contracts. In other trials WebVibe in both Boolean and vector-based form has been made available as a front end to internet search engines.

PROJECT REFERENCES
Korfhage, R.R. 1997. Information Storage and Retrieval. John Wiley & Sons, New York
Morse, E., Lewis, M., Korfhage, R., and Olsen, K. (1998). Evaluation of text, numeric and graphical presentations for information retrieval interfaces: User preference and task performance measures. Proceedings of the 1998 IEEE International Conference on Systems, Man, and Cybernetics, Oct 12-14, San Diego, CA, 1026-1031.
Morse, E. & Lewis, M. (1997). Why information retrieval visualizations sometimes fail, Proceeding of the 1997 IEEE International Conference on Systems, Man, and Cybernetics, Oct. 12-15, Orlando, FL, 1680-1685.

AREA BACKGROUND

Visual interfaces for information retrieval and browsing can be categorized as reference point systems, map displays, or 3-dimensional systems. Reported reference point systems include:

Table 1: Reference Point Visualizations

Visualization System	Reference	Tested?
Component Scale Drawing	Crouch & Korfhage 1990	yes
Cougar	Hearst 1994	no
GUIDO	Nuchprayoon 1996	yes
InfoCrystal	Spoerri 1993	no
SIRRA	Aalbersberg 1995	no
Space	Newby 1992	yes
TileBars	Hearst 1995	yes
VIBE	Olsen et al 1992	yes
WebVIBE	Morse & Lewis 1997	yes

These interfaces can be roughly categorized by their visual representation of similarity as:

Venn diagram: Cougar and InfoCrystal Icon list: SIRRA and TileBars "Spring": VIBE, WebVIBE, and Space Scatterplot: GUIDO, C S Drawing

AREA REFERENCES

Crouch, D.B. and R.R. Korfhage. 1990. The use of visual representations in information retrieval applications. In T. Ichikawa, E. Jungert, & R.R. Korfhage (Eds.), Visual Languages and Applications, pp. 305-326, New York: Plenum

Hearst, M.A. 1994. Using categories to provide context for full-tex retrieval results. In Proceedings of the RIAO '94, New York.

Hearst, M.A. 1995. TileBars: visualization of term distribution information in full text information access. CHI 95 Proceedings, 213-220

Korfhage, R.R. 1997. Information Storage and Retrieval. John Wiley & Sons, New York.

Morse, E. & Lewis, M. (1997). Why information retrieval visualizations sometimes fail, in

Proceeding of the 1997 IEEE International Conference on Systems, Man, and Cybernetics, Oct. 12-15, Orlando,FL, 1680-1685

Newby, G.B. 1992. An investigation of the role of navigation for information retrieval. Proceedings of ASIS 92, 20-25

Nuchprayoon, A. & Korfhage, R. R. (1994). GUIDO, A Visual Tool for Retrieving Documents. In Proceedings 1994 IEEE Computer Society Workshop on Visual Languages, St. Louis, 64-71.

Olsen, K.A., R.R. Korfhage, M.B. Spring, K.M. Sochats, and J.G. Williams. 1993. Visualization of a document collection: The VIBE system. Information Processing and Management. 29(1): 69-81.

Spoerri, A. (1993). InfoCrystal: A Visual Tool for Information Retrieval. In Proceedings Visualization '93, San Jose, CA, 150-157.