zaphiris et al. - 2004 - exploring the use of information visualization for digital libraries - new...

7/30/2019 Zaphiris Et Al. - 2004 - Exploring the Use of Information Visualization for Digital Libraries - New Review of Informat

1/20

EXPLORING THE USE OF

INFORMATION VISUALIZATION FOR

DIGITAL LIBRARIES

Panayiotis Zaphiris, Kulvinder Gill, Terry H.-Y. Ma,Stephanie Wilson and Helen Petrie

This paper presents the findings from a study conducted to explore the current and

future use of information visualization (IV) for digital libraries (DL). A series of query

techniques and participatory design workshops were employed with the goal of actively

involving users in identifying the key characteristics and applications of IV for DLs. The

methodology employed, key findings and overall conclusions are presented and

discussed.

Introduction

The work reported in this paper is the outcome of a study we conducted to

investigate the current use of information visualization (IV) techniques by the

joint information systems committee (JISC) services in the UK and future possible

uses of such techniques in digital library (DL) applications. One of the key

objectives of this project was to actively involve users (through the use of

Participatory Design) in the development of low fidelity prototypes that illustrate

the use of IV for web-based, end-user oriented DL services.

The paper first presents the theoretical foundations for IV and DLs. It then

explores the link between these two fields and finally describes the methodology

employed, the results obtained and the overall conclusions and recommenda-

tions of the project.

Theoretical Foundations

In recent years a number of influential IV techniques have been developed

and found application in different domains (e.g. starfield displays (Albert and

Shneiderman 1994a), Bead (Chalmers and Chitson 1992) and self-organizing maps

(Lagus et al. 1996).

New challenges in IV emerged with the increasing use of the World Wide

Web. Many visualizations have been developed for the purpose of browsing large

collections of online documents, or large data sets (e.g. FilmFinder (Albert and

Shneiderman 1994b)). In addition, the use of different information visualization

techniques for navigation (Spence 2001) have recently been the focus of anumber of studies (Zaphiris et al. 2002; Bederson and Schneiderman 2003).


2/20

What is Information Visualization (IV)?

Card et al. (1999) define visualization as:

The use of computer-supported, interactive, visual representations of data

to amplify cognition. (Card et al. 1999)

They define IV as

The use of computer-supported, interactive, visual representations of

abstract data to amplify cognition. (Card et al. 1999)

The main goals of IV are discovery, decision making, and explanation. IV is

useful to the extent that it increases our ability to perform these and other

cognitive activities. In visualization, abstractions are based on physical space,

whilst non-physical information such as financial data, business information,

collections of documents, and abstract conceptions may also benefit from being

presented in a visual form.

In the information age, where the volume and use of online information is

increasing, there is a need for new forms of presentation and manipulation of

electronic data. Tufte (2001) described two fundamental rules for visual display:

. Maximize the data-ink ratio, i.e. every drop of ink, or pixel on your screen,

ought to be information bearing. Anything that appears simply for decoration

should be removed.

. Maximize information density, i.e. prefer displays with more rather than

less information.

Tufte (2001) also observes that maps are a superior means of applying

these rules, as they convey more information per square unit of display area than

other presentation techniques.

It is thought that visualizing information by representing the semantic

relationships (through metaphors like spatial proximity and visual links), can

facilitate the development and application of the users cognitive map of the

information space (Egan 1988; Vicente and Willeges 1988; Hook et al. 1996). Thismakes the task of searching and browsing for information similar to that of real

world navigation. For example, map overviews (as an aid to hypertext navigation)

have been found to benefit users (Vicente and Willeges 1988; Sein et al. 1993;

Stanney and Salvendy 1995). Furthermore, the use of virtual environments takes

the notion of visualization one step further by providing full, real-time

interactivity and the ability to view the relationships between objects from an

unlimited number of perspectives (Cribbin 2003).

Information Visualization (IV) and Digital Libraries (DLs)

52 PANAYIOTIS ZAPHIRIS ET AL.


3/20

lingual collections of documents that are distributed and accessed worldwide.

Given the fact that they are becoming the main repository of mankinds

knowledge, the design of user-friendly interfaces to access, understand, and

manage DL content has become an active and challenging field of study (Fox and

Urs 2002).

With the ability to share worldwide collections of information, DLs have

become one of the common means for sharing and disseminating information by

individuals or groups that select, organize and catalogue large numbers of

documents.

According to Borner and Chen (2002), our primary means of accessing DLs

today are search engines that retrieve very large numbers of documents relevant

to our search terms. However, search interfaces lack the ability to support

information exploration, making it increasingly difficult for scientists andpractitioners to gain a macro view of DLs, to locate germane resources, to

monitor the evolution of their own and other knowledge domain, and to trace

the influence of theories within and across domains.

The increasing availability of online information makes it necessary for

librarians and information professionals to efficiently and effectively catalogue

the large amount of documents produced (Borner and Chen 2002). There is a

need for new tools that will assist scientists and researchers to identify and

manage all these useful information resources.

The Application of Information Visualization (IV) to DigitalLibraries (DLs)

In applying IV to DL interfaces, the aim is to shift the users mental load

from slow reading to faster perceptual processes such as visual pattern

recognition.

According to Borner and Chen (2002), there are three common usage

scenarios for visual interfaces to DLs:

1. To support the identification of the composition of a retrieval result,

understand the interrelation of retrieved documents to one another, and

refine a search.

2. To gain an overview of the coverage of a DL and to facilitate browsing.

3. To visualize user interaction data in relation to available documents in order

to evaluate and improve DL usage.

Furthermore, Borner and Chen (2002) give suggestions on effective design

and research directions in information visualization and on how visualizationcould assist DLs:

INFORMATION VISUAL 53


4/20

. Provide new means to interact with data, for example, to provide an overview

of what is covered by a DL, enable users to filter out relevant documents

and to examine relationships among those documents.

. Good labeling to ensure that the selection of meaningful words and their

display would not be over plotted.

. To provide multiple perspectives to one data source.

. To support searching of text, image, video, spatial data.

. Personal baskets for users to store previously selected document sets.

. Usability and usefulness studies are needed to improve interfaces and to

specify what does and does not work.

. Strong collaboration among librarians and programmers will help to improve

the design usability of interfaces considerably.

When designing DLs, designing meaningful overviews so that patterns canbe easily recognized, creating comprehensible interfaces to specify what they

want, and providing effective displays of search results are real challenges. The

visual information seeking mantra by Ben Shneiderman:

Overview first, zoom and filter, then details on demand. (Shneiderman

1994)

This provides a good starting point. By interpreting this mantra, a remarkably

diverse set of research and commercial interfaces (Card et al. 1999; Shneiderman

1998; Reed and Heller 1997) demonstrating the use of IV in DLs has been

produced.

Digital Library (DL) Tasks and Information Visualization (IV)

Three key tasks in DLs are searching, navigating and browsing. We describe

below the link of IV with these tasks:

Searching

Search results from DLs are often displayed as a textual list, with 10/20

items per page. Shneiderman et al. (1999) proposed a framework that increases

clarity and user control while it reduces inconsistencies in text-search user

interfaces. This allows designers of specific systems to offer a variety of features in

an orderly and consistent way. Users begin the search process by considering

their information needs and clarifying their search goals, after which they are

ready to employ a computer-based system for the four phases: formulation (what

happens before the user starts a search); action (starting the search); review of

results (what the user sees resulting from the search); and refinement (whathappens after review of results and before going back to formulation).



5/20

Shniederman et al. (1999) suggested the use of a two-dimensional display

with continuous variables to view several thousands search results at once. It has

been applied to a digital video library, a legal information system and to a

technical library using the ACM Computing Classification System.

Navigation

In terms of navigation, research in information visualizations (Card et al.

1999; Shneiderman 1998; Bederson and Shneiderman 2003) produced a

remarkably diverse set of interfaces.

Two important strategies employed to achieve efficient navigation are:

1. Two-dimensional visualizations.

Two-dimensional visualizations with meaningful axes have been an effectivevisualization strategy because thousands of items can be shown at once.

Typical axes make use of continuous variables (e.g. relevance or publication

date) or categorical variables (e.g. language or name of journal). The labels on

such axes are cues to searchers and may contain valuable information if they

are ordinal variables.

2. Browsers for hierarchical data sets.

Hierarchical structures are common nowadays and very often successful, (e.g.

Yahoo!). Hierarchies have the potential to reduce complexity by organizing

related information into comprehensible structures. Hierarchies are typicallyshown by textual lists of one level at a time, by node-link diagrams, cone trees

(Hearst and Karadi 1997) or an outliner in which levels and branches can be

expanded and contracted (Nation et al. 1997).

Graphical Interface for Digital Libraries (GRIDL) (Shneiderman et al. 1999)

adopts these two important strategies by combining the use of hierarchical

browse tool with a two-dimensional visualization. This combination preserves

visual overviews and enables users to rapidly comprehend search results.

Browsing

According to Shneiderman (1997), tasks can range from specific fact-finding

to more unstructured open-ended browsing of known databases and exploration

of the availability of information on a topic.

Several techniques have been developed in the last few years focusing on

interfaces that allow users to browse and search collections of information.

Example include: the dynamic queries (Ahlberg and Shneiderman 1996), visual

information seeking (Ahlberg and Shneiderman 1994) and query previews (Doanet al. 1996; North et al. 1996).



6/20

feature a visual representation of a database (typically with a scatterplot or

starfield display), a visual representation of a query using a collection of widgets

(e.g. sliders), and tight coupling between these two components. Users browse

the database by interacting with the interface widgets. Each change produces a

new query, the results of which are immediately and continuously shown in thedisplay, supporting a progressive refinement of the search, continuous reformu-

lation of goals, and visual scanning to identify results. Query previews also deal

with large distributed databases, and use previews of the data to maintain real

time feedback and limit access to the network.

Participatory Design

Participatory design (PD) (often termed the Scandinavian Challenge

(Bjeknes et al. 1987)) refers to a design approach that focuses on the intended

user of the service or product, and advocates the active involvement of users

throughout the design process. User involvement is seen as critical both because

users are the experts in the work practices supported by these technologies and

because users ultimately will be the ones creating new practices in response to

new technologies (Blomberg and Henderson 1990).

Blomberg and Henderson (1990) characterize the PD approach as

advocating three tenets: (a) The goal is to improve the quality of life; (b) The

orientation is collaborative; (c) The process is iterative.We hypothesized that the use of such an approach as a means of

requirement acquisition through prototyping would achieve high quality of data

and would also motivate our participants to share freely their views and opinions

about the use of IV in DLs.

We now present the methodology we followed and our key findings.

Methodology

The main objective of this project was to engage in the development of IVlow-fidelity prototypes. Prototypes were developed for each of the four

categories of Joint Information Systems Committee (JISC) services that were

considered during the project: Portal, Image, Bibliographic and Geospatial

services.

The JISC was established as an advisory committee, working on behalf of

funding councils to provide and support the implementation of Information and

Computing Technologies (ICTs) in further and higher education. They have

already achieved many of their goals by providing expertise, independent advice,

guidance and key resources to help institutions throughout the country deliver ahigh level of service.



7/20

environment providing secure and convenient access to a comprehensive

collection of scholarly and educational material (JISC, 2002a). With this aim it is

essential that all resources are successfully managed and presented in the most

coherent way.

The Information Environment (IE) is composed of a number of services each

falling under one of four key categories:

. Portals (Resource Discovery Services)

. Bibliographic Services

. Geospatial Services

. Digital image libraries

A key concern for JISC is the wide variety of users needs and the usability

of their services:

Users do not all want to access information in the same way but will require

a diverse range of views of resources in order to satisfy their needs (JISC 2002).

IV is considered one of the ways to enhance the usability of services.

Portals (Resource Discovery Services)

Portals are a single point of entry. Characteristic examples of portals include

yahoo!, bbc.co.uk etc. Resource Discovery Service (RDS) improves the current

search engines and offers better search accuracy for the users by using theirprofile information.

Bibliographic services

Bibliographic services contain databases in a form of an organized

collection of information. The database contains descriptive information (citation

and subject headings) for publications, such as books, periodical articles,

videotapes or government documents. The structure of the database generally

consists of the following information:

. Index / includes citation and subject headings, also known as descriptors, for

each publication. Catalyst, the librarys online catalogue is an example of an

index database.

. Abstracted Index / includes the citation, subject headings (descriptors) and

a summary of the content of the publication.

Geospatial services

One of the key geo-spatial services provided by JISC is Digimap, one of the

f f i t l lib i h d l i f ti i th W ld Wid



8/20

static maps stored on their database, but also quickly generate maps on the fly in

relation to the users specific needs.

Digital image librariesThe number of images available on the Web was estimated in 1997 to be

between 10 and 30 million (Eakins and Graham 1999), a figure that we can

assume has vastly increased in the last five years. Digital image libraries are now a

key resource in retriving such images over the Internet.

Experimental Design

Representative users were recruited to participate in PD sessions. In total 22

individuals were recruited to take part in these sessions. Six were present for the

Portal, Images and Bibliographical services and four for the Geospatial service.

Care was taken so that users participated in the design of prototypes for services

in which they had good expertise (e.g. librarians together with students

participated in the Bibliographical session).

All sessions were fully equipped with the necessary stationery i.e. colored

pens, pencils, felt tips, scissors, card, paper, acetates, overhead projector, post-it

notes etc to help the participants in getting their thoughts on paper. The sessions

were also video recorded and photographed.

The structure for each session was as follows:

1. Welcome and introduction: Participants were welcomed and asked to read

and sign a consent form.

2. Introduction to information visualization: A high-level introduction to

information visualization was presented. The introduction also outlined

common misconceptions regarding information visualization.

3. Introduction to the service: An introduction to the particular service was

provided (for instance, a description of a portal along with specific examples

of portal sites).

4. Requirements gathering: The first activity of the session focused on getting

the participants to start thinking about the service that they were looking

at. The technique used here was that of affinity diagramming. Affinity

diagramming is a simple and effective method to gather a number of ideas

that are then organized by grouping related items together (Institute 1996).

A specific scenario was given for each service and the participants were

required to individually brainstorm ideas on post-it notes in terms of what

information and/or functional requirements users have. The participantswere asked to restrict themselves to one idea per post-it note.

Th i i h k d l h f h i i d



9/20

already been placed on the board. This was repeated until all of their ideas

were put on the board.

The facilitator then led the participants in naming the groupings by reading

out all the requirements per category and giving the opportunity to the

participants to reshuffle the requirements to other categories if needed.Participants also highlighted any relationships that existed among the

categories.

Thirty minutes were allocated to this activity.

5. Documentation on information visualization techniques: Samples of

information visualization techniques were given to the participants to

browse for five minutes. The techniques were not discussed with the

participants to prevent any bias.

6. Paper-based prototyping in pairs: The participants were asked to work in pairsto sketch design ideas using some form of information visualization for the

given scenario using the ideas that had been generated in the previous activity.

They were asked to restrict their designs to one of the categories and not

design a whole interface for the entire web site (for example concentrating

on the search function).

The participants were allowed to refer to the documentation provided and

use any or none of the techniques illustrated.

Thirty-five minutes were allocated to this activity but, depending on the

stage reached by the participants, an additional ten minutes could be

granted.

7. Presentation of initial designs: Each pair presented their design explaining how

it worked and where information visualization techniques were applied.

8. Final prototype: To develop the final prototype, each group was asked to

choose one of three options: (a) work in existing pairs on the same prototype;

(b) work together as a whole group developing one prototype to the next level;

(c) work together as a whole group starting afresh and developing a new

prototype.

All groups chose to work together either developing an existing idea and

taking characteristics from other prototypes on board or developing a new

prototype combining methods/techniques or thinking of new ways to

present the information. The groups were given 40 minutes to design their

prototype.

Results

As an illustration of our overall results from this study we present in this

ti l th fi di f th ti i t d i i f th t l



10/20

Portals: Requirements Gathering

The hypothetical scenario setting given to the participants was:

JISC is a public organization that supports further and higher education.

JISC has released a tender to develop a new portal site aimed at students infurther and higher education to provide information and resources on Music. The

tender requires you to: identify the requirements of the site i.e. what users want

in terms of information and functionality.

Figure 1 shows the findings from the affinity diagram exercise. The

participants proposed twelve main categories of functionality/content for this

service (namely: setting, classifieds/second hand instruments, concerts and

tickets, online music, intercultural music, communications, news, further informa-

tion and outside links, learning music, information about artists, musical genres,

search) with a number of elements under each category.The participants also established some relationships among the categories/

items between news (charts), intercultural music (charts across the world) and

online music (download music and audio samples).

Portals: Paper-Based Prototypes

The participants worked in pairs taking the findings of the requirements

gathering into consideration to design an idea by applying some form of

FIGURE 1

Portals affinity diagram.



11/20

Group 1:

The purpose of this design (Figure 2) was to track specific music journals

across the world.

. The dots on the world map represent cities.

. The selection criteria are presented below the map. The user can select

the country they wish to find music journals for, the timeframe (1960 /1970),

the chart level (e.g. top ten), and the Genre (e.g. pop, jazz, folk etc).

. The results are displayed in a color-coded format and symbols on the map.

. It is possible to filter out information and have the information presented

in different views. The user can zoom in and zoom out at a continent and

country level.

. Data between countries can be compared, and viewed using the

other visualization methods such as histograms, bar charts, pie charts etc

by manipulating the criteria.

. A pop-up window appears to view details of selection e.g. top ten charts in

New York.

The design was inspired by two information visualization techniques:

d (D t l 2001) d i D f f d i

FIGURE 2

Portals Group 1.



12/20

represented in the map. Users adjust the widgets to form composite queries and

the map elements are then shaded in response to the query. Zooming interfaces

allow the user to smoothly zoom into the visualization in order to view more

detailed information and to zoom out for overview information.

Group 2:

The purpose of this design (Figure 3) was to demonstrate the sales of

albums and concerts worldwide.

. The user can search for sales or concerts worldwide by selecting the

appropriate tab/link at the top of the page.

. The user can zoom in to continent, country, city, town via the worldwide map.

. The search settings on the right help refine the criteria, as once the userhas selected the desired location they can search by period (1990/1994),



13/20

number of sales, genre and similar music. The combo boxes expand to allow

the user to select/refine the search criteria.

. The results are immediately displayed in the bottom left hand box. The

squares, triangles and circles represent albums, singles and DVD sales,

respectively. The Y axis shows the number of sales and the X axis shows theyear it was published. (The triangles etc are not grouped together because

they are similar but because they have similar number of sales.)

. The labels represent the different groups (genres) and are placed on the

display for easier selection.

. On selecting a label, further information is provided on the right beneath

the search settings.

The design was inspired by the following information visualization

techniques: dynamaps (Dang et al. 2001), zooming, excentric labeling (Feketeand Plaisant 1999), and Shneidermans FilmFinder (Ahlberg and Shneiderman

1994) application. Excentric labeling is a technique whereby objects in the

general neighborhood of the cursor are labeled dynamically. FilmFinder was an

early example of a dynamic query interface. The user is presented with a two-

dimensional scattergram representation of the dataset which they can filter

dynamically using alphaslider widgets and can also zoom in and out.

Group 3:

The purpose of this design (Figure 4) was to learn about music from

different cultures.



14/20

. Selecting a location zooms in to select a city, town.

. Information can also be searched using the search on the top right. If search

is used, the map will zoom in to the selected city.

. Details on the right hand side of the screen are empty until a user selects

a location.. When a user selects a location, e.g. Egypt, the Associated Genres appear in the

top right hand side of the screen.

. On selecting one of the options provided in the Associated Genres window

further details are provided in the pop-up window shown below.

This design was also inspired by the dynamaps (Dang et al. 2001) and

zooming IV techniques.

Portals: Final Prototype

Finally, the participants worked in a group of six to design the final set of

prototypes (Figures 5 and 6).

The first prototype was an enhancement of Group 2s prototype comparing

the number of songs reaching number one.



15/20

. The globe presents an overview of the information; the selection of an area

will zoom in to country, city, and/or town. A user selects two areas e.g.

Australia and U.K to compare the song and the place it appeared in the

chart. The results are displayed in the bar chart.

.

If the user enters the interface via the globe, they can further refine theirsearch using the timeframe on top of the bar chart.

. Alternatively the user may enter their criteria using the search settings which

are presented in the top right hand corner.

. On selecting a particular entry from the bar chart specific information about

that song is presented under the search settings e.g. singer, email, lyrics,

number of weeks in top ten.

. Changing the criteria immediately updates the bar chart.

. The problem outlined by the participants was that too much information

is presented for small screens.


FIGURE 6

Portals Combined Prototype 2.



16/20

The second version was an enhancement of Group 3s concept attempting

to show information on music from different cultures.

. The globe presents an overview of the information; the selection of an area

will zoom in to country, city, and/or town.. On selecting a particular area, the pie chart will show music from that culture

and will illustrate where in the world that music is listened to/produced. The

legend will show the percentage.

. The search settings allow a user to further refine the criteria. For example,

by selecting Egypt as the country, Arabic as the culture and music produced

in 2000, the pie chart will depict all Arabic music across the world.

. Changing the criteria immediately updates the pie chart.


technique: dynamaps (Dang et al. 2001) and zooming.

Conclusions

In our view the key finding of this study is the fact that it has demonstrated

the benefit of involving users and other key stakeholders in the design of

interactive DL services.

Through our study we found that all stakeholders had valuable suggestions

and input to make when it comes to designing IV techniques for DLs.

Key Findings

1. Our participatory design sessions through input from users and other

stakeholders developed representative paper prototypes of IV techniques

and methods that are applicable to specific web-based, end user-oriented

JISC services.

2. Our participatory design results section specifically proposes which IV

techniques have been suggested for which type of JISC service (portals,

images, bibliographic or geo-spatial).

Specific Findings

We noticed that the majority of participants of our focus groups, although

involved in their majority with the development and maintenance of JISC

services, were not familiar with IV. Participants in the geo-spatial focus groups

were more familiar with IV as this area is considered to be closely related to the

discipline of Geospatial information.There were misconceptions with participants thinking that IV is simply

b t 3D d i t l lit S ti i t d th t IV



17/20

The majority of participants thought that visual presentations (e.g. maps)

are useful in assisting their search when using DLs. Both stakeholders and end

users were interested in exploring how the relationships between results could

be display / i.e. how the results generated correlate between each other and

how users could compare the results at a glance. They thought that IV might behelpful in contributing towards this issue.

For geo-spatial libraries, experts in geo-spatial information systems stressed

that the ideal geo-spatial libraries do not only shows the GI data but make it able

to link and correlates the GI data with non-GI data / national service/data (e.g.

economic and social data, census data) in order to assist them in their research.

But in order to do this, it will takes an extensive among of afford by all relevant

agencies (especially owners of the data sets).

Participants of the portal, images and bibliographic focus groups, also

raised the issue of designing applications that show relationships among data,

and these relationships need to be obvious and represented visually.

Zooming was by far the most popular technique applied to the prototypes

followed by dynamaps. Zoom enables users to magnify specific aspects of the

site, whereas dynamaps permits users to manipulate information from a set of

widgets (options) to simultaneously present the data on the interface. The

common ground for both of these techniques is that the user controls the level of

detail that is presented.

Based on the results of the participatory design sessions we can

recommend that the following techniques should be further investigated in

terms of appropriateness, usability and accessibility:

. Portals: dynamaps and zooming.

. Images: graphical interface for DLs, PhotoMesa image browser and

generalized query previews.

. Geospatial: zooming.

. Bibliographical: graphical interface for DLs.

The participatory design method was a cost-effective approach to get

feedback on how IV can be applied to web services. Participants were eager to

provide their views and insights in designing user-friendly IV interfaces.

ACKNOWLEDGEMENTS

This project was funded by the JISC Information Visualisation Foundation Study

grant.

REFERENCES

D i i f i f i



18/20

AHLBERG, C. and SHNEIDERMAN, B. 1994. Visual information seeking: tight coupling of

dynamic query filters with starfield displays. In: ACM CHI94: Human Factors in

Computing Systems . Boston: ACM Press.

ALBERT, C. and SHNEIDERMAN, B. 1994a. Visual information seeking: tight coupling of

dynamic query filters with starfield displays. In: ACM CHI94 Conference . Boston:ACM Press.

*/*/. 1994b. Visual information seeking using the FilmFinder. In: ACM CHI94

Conference . Boston, MA: ACM Press.

BEDERSON, B. and SHNEIDERMAN, B. eds. 2003. The craft of information visualisation . San

Francisco: Morgan Kaufmann.

BJERKNES, G., PELLE, E. and MORTEN K. 1987. Computers and democracy*/a scandinavian

challenge . Aldershot, England: Avebury.

BLOMBERG, J. L. and HENDERSON, A. 1990. Reflections on participatory design: lessons from

the trillium experience. In ACM CHI90: Human Factors in Computing Systems .

Seattle, WA: ACM Press.

BORNER, K., CHEN, C. 2002. Visual interfaces to digital libraries: motivation, utilization, and

socio-technical challenges. In Visual interfaces to digital libraries , edited by K.

Borner and C. Chen. New York, NY: Springer.

CARD, S., MACKINLAY, J. and SHNEIDERMAN, B. 1999. Readings in information visualization:

using vision to think. San Francisco, CA: Morgan Kaufmann Publishers.

CHALMERS, M. and CHITSON, P. 1992. Bead: explorations in information visualisation. In

SIGIR92 . Copenhagen: ACM Press.

CRIBBIN, T. 2003. Web resources on information visualisation , http://www.cribbin.fsnet.

co.uk/infovis/infovis.htm

DOAN, K., PLAISANT, C. and SHNEIDERMAN, B. 1996. Query previews for networked

information services. In Advanced Digital Libraries Conference . Washington: IEEE.

EAKINS, P. and GRAHAM, M. 1999. Content-based Image retrieval: A report to the JISC

Technology Applications Programme , http://www.unn.ac.uk/iidr/research/cbir/

report.html#Heading2

EGAN, D. 1988. Individual differences in human-computer interaction. In Handbook of

human-computer interaction , edited by M. Helander. North-Holland: Elsevier.FEKETE, J.-D. and PLAISANT, C. 1999. Excentric labeling: dynamic neighborhood labeling for

data visualization. In ACM CHI99: Conference on Human factors in Computer

Systems . New York: ACM.

FOX, E. and URS, S. 2002. Digital libraries. Annual Review of Information Science and

Technology 36: 503/89.

HEARST, M. and KARADI, C. 1997. Cate-a-Cone: An interactive interface for specifying

searchers and viewing retrieval results using a large category hierarchy. In 20th

Annual International ACM SIGIR Conference . Philadelphia: ACM Press.

HOOK, K., SJOLINDER, M. and DAHLBACK, N. 1996. Individual differences and navigation inhypermedia. In 8th European Conference on Cognitive Ergonomics (ECCE-8) .

S i



19/20

LAGUS, K., HONKELA, T., KASKI, S. and KOHONEN, T. 1996. Self-organizing maps of document

collections: A new approach to interactive exploration. In Second International

Conference on Knowledge Discovery and Data Mining , edited by E. Simoudis, J.

Han, and U. Fayyad. Menlo Park: AAAI Press.

LESK, M. 1997. Practical digital libraries: books, bytes, and bucks . San Francisco: MorganKaufmann.

NATION, D., PLAISANT, C., MARCHIONINI, G. and KOMLODI, A. 1997. Visualizing websites using a

hierarchical table of contents browser: WebTOC. In: 3rd Conference on Human

Factors and the Web . Basking Ridge, NJ: HFES.

NORTH, C., SHNEIDERMAN, B. and PLAISANT, C. 1996. User controlled overviews of an image

library: a case study of the visible human. In: 1st ACM International Conference

on Digital Libraries . Bethesda, MD: ACM Press.

REED, M. and HELLER, D. 1997. Online library of information visualisation environments ,

http://otal.umd.edu/OliveSEIN, M.K., OLFMAN, L., BOSTROM, R.P. and DAVIS, S. 1993. Visualization ability as a predictor

of user learning success. International Journal of Man-Machine Studies, 39: 599/

620.

SHNEIDERMAN, B. 1997. Designing information-abundant web sites: issues and

recommendations. International Journal Human Computer Studies, 47: 5/29.

*/*/. 1998. Designing the user interface. Strategies for effective human-computer

interaction. 3rd ed. Reading, MA: Addison-Wesley.

SHNEIDERMAN, B., FELDMAN, D., ROSE, A. and GARU, X.F. 1999. Digital library search results

with categorical and hierarchical axes. In DL-00: 5th ACM Digital Library

Conference . San Antonio: ACM Press.

SPENCE, R. 2001. Information visualization. West Sussex: ACM Press.

STANNEY, K. and SALVENDY, G. 1995. Information visualization: assisting low-spatial

individuals with information access tasks through the use of visual mediators.

Ergonomics , 38(6): 1184/98.

TUFTE, E. 2001. Envisioning information . Cheshire, Connecticut: Graphics Press.

VICENTE, K. and WILLEGES, R. 1988. Accommodating individual differences in searching a

hierarchical file system. International Journal of Man-Machine studies , 29: 647/

68.

ZAPHIRIS, P., SHNEIDERMAN, B. and NORMAN, K.L. 2002. Expandable indexes versus sequential

menus for searching hierarchies on the World Wide Web. Behaviour and

Information Technology, 21(3): 185/201.

Dr Panayiotis Zaphiris, Centre for HCI Design, City University, London, EC1V

0HB, UK. Tel: '44-(0) 20-7040-8168; Fax: '44-(0) 20-7040-8859. E-mail:

[email protected]



20/20

zaphiris et al. - 2004 - exploring the use of information visualization for digital libraries - new...

Documents