Supporting University Research and Administration via

Interactive Visual Exploration of Bibliographic Data

Kostiantyn Kucher

1 a

and Andreas Kerren

1,2 b

Department of Science and Technology, Link

oping University, Norrk

oping, Sweden

Department of Computer Science and Media Technology, Linnaeus University, V

axj

o, Sweden

Keywords:

Bibliographic Data, Bibliometric Analysis, Data Curation, Co-Authorship Networks, Library and Information

Science, Publication Data Visualization, Scholarly Data Visualization, Information Visualization.

Abstract:

Bibliographic data and bibliometric analyses play an important role in the professional life of academic re-

searchers, and the quality of the respective publication records is essential for establishing the big picture of

the relationships between particular publications, their authors and afﬁliations, or further data facets associated

with publications. In this paper, we report on the design and outcomes of an interactive visual data exploration

project conducted within the scope of a university with the goal of gaining overview of the university publi-

cation data. The project has been carried out by information visualization researchers in collaboration with

several groups of stakeholders, including the university library and administration staff. We describe the de-

sign considerations, the resulting interactive visual interface, and the feedback received from the stakeholders

with respect to the tool functionality and the insights discovered in the bibliographic data.

1 INTRODUCTION

Scientiﬁc publications constitute an important part of

the research output produced by the majority of aca-

demic researchers. While the views and policies re-

garding the publishing forms and bibliometric mod-

els may differ across individual researchers and in-

stitutions, the importance of the bibliographic data

quality would arguably be accepted by most. Reli-

able publication data provides a rich source of infor-

mation not only on the particular publications them-

selves, but also various derived data such as the pub-

lication statistics for individual researchers or groups,

co-authorship networks and collaborations at various

levels, and many more. The respective insights might

be sought after with manual exploration of biblio-

graphic data, bibliometric and scientometric analy-

ses (Small, 2006), and interactive visual approaches

(Federico et al., 2017; Liu et al., 2018).

While the related work includes impressive ex-

amples of advanced computational and interactive

analyses—and furthermore, some commercial solu-

tions are available—application of such approaches

is not always feasible or does not always address the

needs and preferences of particular stakeholders. This

https://orcid.org/0000-0002-1907-7820

https://orcid.org/0000-0002-0519-2537

leaves room for design and implementation of custom

solutions tailored for particular data, users, and tasks.

In this paper, we report on the design/application

study initiated by the university library and university

administration staff and resulting in an interactive vi-

sualization tool for publication data available within

our home university (see Figure 1). The contributions

of this paper are the following:

• domain task characterization for several groups of

stakeholders;

• analysis of the data-centric and user-centric de-

sign requirements;

• proposed design of the backend and interactive

visual interface for publication data exploration

from the perspectives of data hierarchy, heteroge-

neous network, and further facets; and

• discussion of the feedback from two groups of

stakeholders, namely, the university library and

university administration staff.

Section 2 provides an overview of the related work

in library and information science as well as the in-

teractive approaches developed mainly within the vi-

sualization research community. Then we focus on

the analysis of stakeholders and their requirements for

the planned approach in Section 3. Since the chal-

lenges to be addressed can be roughly divided into

248

Kucher, K. and Kerren, A.

Supporting University Research and Administration via Interactive Visual Exploration of Bibliographic Data.

DOI: 10.5220/0011806900003417

In Proceedings of the 18th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (VISIGRAPP 2023) - Volume 3: IVAPP, pages

248-255

ISBN: 978-989-758-634-7; ISSN: 2184-4321

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

University

Bibliographic Data

Management Platform

Supplementary

Publication Venue

Rankings

Data Retrieval and

Processing Module

Supplementary Data

Importing Module

Web Application Interactive Visual

Interface

Users

University

Librarian

University

Manager

University

Researcher

External

User

Publication

Search/Filtering

Hierarchy

View

Network

View

Data Facet

Views

Figure 1: Overview of our proposed interactive visual approach for bibliographic data exploration.

two groups, we ﬁrst discuss the data-related concerns

in Section 4 and then the interactive visual interface

design in Section 5. We report on the stakeholders’

feedback and discuss further concerns in Section 6,

and conclude this paper with Section 7.

2 RELATED WORK

Application of computational methods involving sta-

tistical, temporal, and graph/network-based models

has a long history in bibliometrics and scientomet-

rics. The researchers in library and information sci-

ence also make use of visualization, especially for

graphs/networks constructed from publication data.

For example, Small describes computational and vi-

sual approaches for (co-)citation networks, which can

be eventually applied to predict emerging areas of

growth across scientiﬁc disciplines and ﬁelds (Small,

1999; Small, 2006). While such studies, conducted

and published primarily by library and information

science researchers, mainly make use of traditional

statistical charts and static node-link diagram rep-

resentations, Salaba and Mer

cun report the results

of an interesting user study that compares a tradi-

tional faceted user interface for bibliographic data ex-

ploration with several interactive visualization tech-

niques less widely used in the community, such as

a sunburst diagram (Salaba and Mer

cun, 2020), with

encouraging results for particular user tasks.

Within the visualization community, bibliographic

data and bibliometric analyses have also attracted the

researchers’ attention for years. The variety of stud-

ies focusing on publication data are discussed in sev-

eral existing surveys (Federico et al., 2017; Liu et al.,

2018), for instance, the CiteSpace tool (Chen, 2006).

Some of the noteworthy recent contributions in this

ﬁeld include, among others, GRAM by Burd et al., an

approach that generates interactive map-like represen-

tations of aggregated research topic data (Burd et al.,

2018). GRAM does not directly rely on bibliographic

records, though, as it uses self-reported research top-

ics from Google Scholar proﬁles. One of the goals of

the approach by Burd et al. is to facilitate the univer-

sity administration policy-making regarding research

strategies and resources. This is related to the goals

of our project, too; however, the data sources and user

tasks are quite different, leading to different design

choices for the visual representations and interactions.

With respect to the focus on the network data ex-

traction and representation for scientiﬁc publications,

OLGAVis (Jo et al., 2021) provides OLAP function-

ality and node-link visual representation. Some of

the other existing approaches focus on the temporal

perspective, typically with the goal of representing

the publication data from a particular conference over

time, e.g., CiteRivers (Heimerl et al., 2016) or VIS-

tory (Zeng et al., 2021). VisualBib (Dattolo and Cor-

batto, 2022) provides a rich set of bibliography man-

agement functionality in addition to exploration capa-

bilities. In contrast to these approaches, ReviewerNet

(Salinas et al., 2020) shifts the focus to individual re-

searchers rather than venues or publication records.

The study by Rosenthal et al. (Rosenthal et al., 2019)

is one of the most relevant to our work with respect to

the underlying motivation and stakeholders, although

their focus is entirely on the temporal perspective.

3 DESIGN REQUIREMENTS

The overall process of design and realization of

this project ﬁts the general framework of de-

sign/application studies as discussed by the visual-

ization community (Sedlmair et al., 2012). The re-

quirements for this work were motivated and com-

municated by the representatives of several groups of

stakeholders within the university over the course of

months via email and in-person meetings. Table 1

provides an overview of the main groups of stakehold-

ers. The representatives of the ﬁrst two groups were

directly involved in the discussions regarding the de-

Supporting University Research and Administration via Interactive Visual Exploration of Bibliographic Data

249

Table 1: The main groups of stakeholders involved or considered in our project, and their respective needs.

Stakeholder group Needs to be addressed by the interactive tool

University librarians (1) Gain overview of university-wide bibliometric results; (2) identify issues with bibliographic

data

University administration (1) Gain overview of university-wide or more detailed bibliometric results; (2) assess and compare

the publication output at various levels within the university; (3) identify productive existing and

potential collaborations at the internal, national, and international levels

University researchers (1) Gain overview of university-wide or more detailed bibliometric results; (2) identify potential

collaborations and research opportunities at various levels

Potential external users (1) Gain overview of university-wide bibliometric results; (2) identify particular publications, re-

searchers, groups, or projects at the university for potential collaborations, supervision, expert

duties, outreach activities, etc.

Table 2: The main design requirements established for our project.

R1 The interactive tool shall retrieve and use the publication data from the existing bibliographic data management platform

DiVA (M

uller et al., 2003) used at the university

R2 The interactive tool should support data augmentation with respect to the publication venue ranking according to the

speciﬁc ranking data used at the university and available at the national level

R3 The interactive tool should allow for the potential quality issues in the existing publication data, such as missing or mis-

spelled afﬁliation titles, for instance

R4 The interactive tool shall be available as a web-based application (and it will be primarily used from desktop/laptop

computers rather than other device classes)

R5 The interactive tool shall provide the bibliometric information based on the existing publication data, including the tempo-

ral overview, top publication venues, research subjects and disciplines, keywords, and external collaborations mentioned

within the publication metadata entries

R6 The interactive tool shall support browsing, search, and ﬁltering across the publication data with respect to the afﬁliations

mentioned within the publication metadata entries, including the support for afﬁliation hierarchies, when possible

R7 The interactive tool should support the exploration of relationships between publication entries, authors/editors, and afﬁli-

ations based on the existing publication data

R8 The interactive tool should provide its functionality to the users with limited training/instruction required (and the docu-

mentation on the main modules, representations, and interactions should be provided within the tool)

sign and implementation of the proposed approach,

and the authors of this work represent the third group

themselves. The fourth group represents further po-

tential end-users, such as students or external actors.

Based on these considerations, the main design re-

quirements listed in Table 2 could be established for

our project. We should mention that the stakehold-

ers (besides ourselves as the authors of the tool) did

not initially provide any hard requirements with re-

spect to the visual representations or interactions to

be included/excluded from the user interface, but the

last requirement on the list captures the expectation

for techniques that would not be overwhelming for

the users without extensive training or background in

information visualization (B

orner et al., 2016; Rus-

sell, 2016). Later during the project development, the

stakeholders actually expressed a requirement for one

particular visualization technique to be added—this

will be discussed below in Section 5.3.

4 DATA-RELATED CONCERNS

As described above, a considerable number of re-

quirements and constraints for the interactive tool

concern the data retrieval, storage, and processing

functionality, since the project aims to make use of

the existing bibliographic data management platform

DiVA (M

uller et al., 2003) used at our university. The

respective platform has been in active use for many

years, which explains the peculiarities and quirks of

the data schema and export formats provided by it.

As presented in Figure 1, the backend consists of

several components, starting with the database client

module. The modules for retrieving and importing

the main publication data sets as well as publication

venue rankings are designed to be either launched as

one-off scripts, or set up to be launched periodically.

In order to accommodate the design requirements

discussed above, the publication entries are not sufﬁ-

cient on their own. The backend module thus extracts

the information about the authors of the publications,

the editors (in case at least one of them is detected to

be the staff member at our university), and the respec-

tive afﬁliations. Furthermore, the information about

publication venues (and their rankings, if available),

publishers, funders, keywords, research subjects, and

disciplines is extracted from the bibliographic records

and stored in the DB to be used by the frontend.

IVAPP 2023 - 14th International Conference on Information Visualization Theory and Applications

250

We should note that even when dealing with the

data within a single university, our tool had to ad-

dress the challenges of matching and disambiguating

author/editor entries, inconsistent afﬁliation records,

and further data quality issues. While the platform

used by our university encourages the researchers to

specify their local account details within the publi-

cation records, it does not enforce a consistent way

to specify the external collaborators and their afﬁli-

ations. This is beneﬁcial in some cases (too much

validation and restriction would probably be cumber-

some and inﬂexible in some scenarios), but for the

purposes of data extraction, analysis, and visual ex-

ploration, this results in additional challenges. For

example, to make use of the external collaborations

data, the external afﬁliation strings are parsed by our

backend using the coco (country converter) library

(Stadler, 2017) and additional heuristics.

Finally, on importing and processing the publica-

tion data sets nightly, the backend prepares and caches

the JSON representation of the processed data set to

be used by the Flask web application and served to

the interactive visual interface, as discussed next.

5 VISUAL INTERFACE

The frontend of our tool is implemented in JavaScript

using D3 and further libraries mentioned below to

provide a rich custom user interface (see Figure 2)

with multiple coordinated views (Roberts, 2007).

The top panel includes the text search and the ﬁlter

controls (cf. Figure 2(a)). The search query is com-

pared to multiple text-based ﬁelds in the data entries,

e.g., publication venues or afﬁliation titles. Some

of the ﬁlters can be adjusted only (e.g., the tempo-

ral ﬁlter) or completely removed (e.g., the ﬁlters as-

sociated with a particular author, etc.). The motiva-

tion for including the permanent, non-removable ﬁl-

ters and for their default values lies with the sugges-

tions and requirements expressed by our stakehold-

ers: for instance, the optional ﬁlter for selecting only

the publications with explicit funding notes was re-

quested by the university administration in order to

focus on the respective data. The ﬁltering eventually

affects the underlying set of the currently displayed

publication entries (as the rest of information such as

authors and afﬁliations is linked to particular publica-

tions) and thus triggers the updates of the other views

discussed below.

5.1 Hierarchical Data View

The left panel displayed in Figure 2(b) represents

the hierarchical data extracted from the publications:

publication entries are nested under persons (au-

thors/editors), persons are nested under afﬁliations

(institutions/departments or external afﬁliations), and

afﬁliations are nested under aggregate afﬁliations

(e.g., external afﬁliations are grouped into countries

and world regions). Thus, a single publication record

from the DiVA platform might be represented by sev-

eral entries in the hierarchy view, being nested under

several co-authors or co-editors, for instance.

Color coding is used sparingly in this part of the

GUI, with blue color used to indicate the entries re-

lated to the expanded network view nodes (see be-

low), and gray color used to indicate the publication

entries with editor rather than author contributions

from the respective persons. The person entries re-

lated to the home university are additionally marked.

Clicking on a panel header in the hierarchy

view (e.g., for the aggregate entries “Home univer-

sity” and “External collaborations” visible by de-

fault) folds/unfolds the corresponding nested hierar-

chy data, thus allowing the user to dive deeper into

details, if desired. Hovering over an icon prepended

to each hierarchy entry title reveals a tooltip with ad-

ditional details. The details are also displayed when

hovering over a badge label appended to the title. The

respective numerical label represents the number of

non-ﬁltered children nodes for the respective hierar-

chy node, such as the number of persons for an afﬁl-

iation. The next control element included in the hier-

archy entry node is a ﬁlter button (e.g., to only display

publications with a speciﬁc author or afﬁliation). Ad-

ditionally, clicking on a button with four arrows will

trigger a node expansion + highlighting update in the

network view (see the next subsection).

For publication entries in particular, a link to the

respective page in the DiVA platform is displayed as

well as the publication venue ranking, if available (cf.

R2 in Section 3). Finally, full details about a publica-

tion are presented in a dialog window when clicking

on a button with an information icon.

5.2 Network Data View

The central panel displayed in Figure 2(c) provides

a heterogeneous network view comprising node rep-

resentations of publications, persons, afﬁliations, and

aggregate afﬁliations. The person and afﬁliation

nodes related to the home university are additionally

marked. In contrast to the hierarchy view, the data

used for the network does not include duplicate nested

Supporting University Research and Administration via Interactive Visual Exploration of Bibliographic Data

251

b c d

Figure 2: The interactive visual interface of our bibliographic data exploration tool: (a) the search and ﬁlter controls; (b) the

hierarchical data view; (c) the network view; and (d) the data facet views. Here, in addition to the ﬁlters applied by default

(which affect all of the views), the user has expanded the network node corresponding to the ISOVIS research group, and

hovered the mouse pointer over the node corresponding to Kerren, Andreas in the network view, triggering the tooltip.

entries—instead, the relationships between various

nodes are represented by edges (lines), e.g., a sin-

gle publication node can be connected to several co-

authors or co-editors. Most of the contents of the net-

work are folded (hidden) initially and can be revealed

through interactions. This decision was inspired by

the egocentric network exploration approach (Fisher,

2005) in order to allow the users focus on the parts of

the network most relevant to their current focus; fur-

thermore, representing the complete network of thou-

sands of nodes (publications, authors, etc.) and edges

is not always feasible with a node-link diagram.

The network view is implemented using the yFiles

for HTML library (Wiese et al., 2004), which supports

a number of layout algorithms and interactions while

providing a high degree of customizability and perfor-

mance. The choice of the layout currently used in our

tool was driven by the underlying data and expected

interactions: while the set of nodes and edges visi-

ble on accessing the tool initially is quite predictable

(top-level aggregate nodes), an arbitrary subset of the

nodes and edges might eventually be displayed, based

on the user’s actions. Currently, the organic layout ap-

proach is used with yFiles, with further enhancements

regarding the identiﬁcation and radial layout for star

substructures, as well as organic edge routing.

The network view supports panning, zooming,

node tooltips on hover, and node highlight on click-

ing: the node itself is highlighted with orange-red,

and its currently visible network connections (edges

and nodes) are highlighted in yellow. This function-

ality is useful to trace particular relations at a glance.

Furthermore, right-clicking a node creates a new pub-

lication ﬁlter (e.g., an afﬁliation ﬁlter in case of an

afﬁliation node).

Finally, double-clicking on a network node will

expand it. This action will reveal the edges and nodes

connected to the double-clicked node, if they are not

displayed already. For example, double-clicking on

a publication node will reveal the related persons as

well as afﬁliations. The network layout will adapt ac-

cordingly. Expanded nodes are highlighted in blue,

and they are not affected by ﬁltering applied in the

user interface, effectively “pinning” them (not with

respect to the spatial position, though). To keep track

of the currently expanded nodes, the corresponding

labels are displayed below the central panel.

Further examples of the outcomes of network

view interactions are demonstrated in Figure 3. Here,

the user has identiﬁed a publication of interest (e.g.,

via the hierarchy or data facet views) and chosen

to expand and highlight the respective node (Chatz-

imparmpas et al., 2020). The adjacent nodes are

then displayed in the network view, meaning au-

thors/editors for the respective publication. The user

decides to focus on one of the persons not afﬁliated

with the home university, Rossi, Fabrice, and expands

the respective network node. This leads to the ap-

pearance of the adjacent nodes, i.e., publications and

afﬁliations for the respective person. No further pub-

lication nodes appear, though, indicating that the re-

spective external collaborator has not contributed to

any other publication in the current data set (given

the current set of ﬁlters); however, an additional ex-

IVAPP 2023 - 14th International Conference on Information Visualization Theory and Applications

252

Figure 3: Example of interactions within the network view.

ternal afﬁliation node appears. The user decides to

expand and highlight the respective node Universit

Paris Dauphine, France. The results of this interac-

tion are demonstrated in Figure 3: the adjacent aggre-

gate afﬁliation node France appears in the network,

and the layout is automatically adjusted to accommo-

date the edges between the afﬁliation, aggregate node

(country), aggregate node (region), and top-level ag-

gregate node for external collaborations.

5.3 Data Facet Views

The right panel displayed in Figure 2(d) comprises

several views related to speciﬁc aspects or attributes

in the publication metadata, most of which are folded

by default to save space and avoid overwhelming the

user. Hovering over most of the elements in the right

panel will reveal a tooltip with additional details, and

buttons are also available to create corresponding ﬁl-

ters (e.g., to only display publications with a speciﬁc

keyword or research subject).

The temporal view provides a slider for ﬁltering

the visible data based on the publication year. It also

includes a bar chart representing the total number and

the currently displayed number (affected by the cur-

rently applied ﬁlters) of publications by year.

The list of recent publications is located below to

provide the user with a simple way to browse the lat-

est publications within the currently displayed data

subset without the need to navigate through the af-

ﬁliations or authors.

The map view provides a simple choropleth rep-

resentation based on the world region data extracted

from the external afﬁliations listed for the publication

records. The intensity of the color used for each world

region is proportional to the number of the currently

visible publications (affected by the applied ﬁlters).

Under the map view, the corresponding information

about world regions is additionally represented with a

list ordered by the respective numerical value.

The rest of the views in the right panel represent

sorted lists of external afﬁliations, publication key-

words and venues, etc., based on the number of the

currently visible publications (affected by the applied

ﬁlters). These lists allow the user to quickly browse

through the top publication venues, for instance.

The headers for these lists also include buttons

for an additional functionality that was explicitly re-

quested by our collaborators from the university ad-

ministration. They asked for the option to generate a

word cloud (Vi

egas and Wattenberg, 2008) for the re-

spective set of weighted labels representing top publi-

cation venues, funders, etc. for the currently displayed

subset of publications. The motivation for this request

was related to the need to export their ﬁndings from

the exploration or speciﬁc analyses within the tool in a

form that would be suitable for further dissemination

and presentation purposes, e.g., as a ﬁgure included

in presentation slides for the use within or outside of

the university. While we mentioned the existing con-

cerns raised within the visualization community in re-

lation to word clouds (Vi

egas and Wattenberg, 2008;

Felix et al., 2018), our collaborators saw this tech-

nique as familiar, suitable for their needs, and aes-

thetically pleasing. Thus, we implemented this tech-

nique, while limiting the max number of entries for

each respective list, allowing for horizontal and ver-

tical alignment only, and providing a dialog with the

options to exclude the particular elements and to edit

the respective text labels. The latter functionality was

motivated by the nature of the underlying data, which

might result in very long titles of journals or funding

agencies, for instance, and result in deteriorated qual-

ity of the layout.

6 DISCUSSION

In this section, we report on the feedback received

from several groups of stakeholders throughout the

project, as well as lessons learned, limitations, and

considerations for future work.

6.1 Feedback from Stakeholders

As part of our communication with the colleagues

from the university library, early on we managed to

ﬁnd common ground and the understanding that the

interactive visual tool would be complementary and

would not be designed to replace the existing bib-

liographic data management platform DiVA (M

uller

et al., 2003) used at our university, which was ben-

eﬁcial for the project. The librarians were also very

helpful with respect to clarifying the particular data

format peculiarities found in DiVA, and were eager to

investigate the bibliographic data quality issues iden-

tiﬁed with both the backend processing and interac-

tive exploration within our tool.

Supporting University Research and Administration via Interactive Visual Exploration of Bibliographic Data

253

One interesting example was related to a biblio-

graphic entry that posed issues for the CSV parser:

on closer inspection, we discovered that the respective

publication author managed to use the rich text editor

within DiVA to upload a screenshot of their paper’s ti-

tle page instead of providing the abstract as text. The

colleagues from the university library also used our

tool to identify inconsistent spellings and typos even

for internal afﬁliations and authors/editors, especially

for older publication entries, which is facilitated by

browsing the hierarchical data view.

One particular suggestion that we received (and

implemented) from the librarians was related to the

network exploration: while navigating through the

network and hierarchy views, they discovered that it

could become cumbersome to look up the network

nodes to fold within the network view; thus, the con-

trols for folding the nodes directly from the labels sit-

uated below the network view were added.

Our colleagues from the university administration

also demonstrated interest and provided encouraging

feedback for the design and functionality offered by

our tool. Some particular adjustments or interactions

that they requested during the project included, for in-

stance, the search by research subject; representation

of local research groups and projects as afﬁliations

within the hierarchy and network views; ﬁltering the

publications by the explicit funding status; and also

the word cloud functionality, with the ability to ex-

clude or adjust the particular items that could be con-

sidered noise.

Some of the interesting ﬁndings discovered by us-

ing our tool were related to the role of particular re-

search funding agencies in the available publication

data for our university. Further comments made by

the university administration representatives included

the notes about the inconsistently speciﬁed afﬁliation

titles/abbreviations (as speciﬁed by the respective au-

thors) and the potential way to address this issue by

integrating our tool further with the internal univer-

sity administration IT systems, e.g., in order to re-

trieve more accurate data about the ongoing projects,

groups, etc. and the respective staff allocation.

6.2 Limitations and Future Work

From the point of view of the data availability and

backend aspects of our project, we should mention

the potential inaccuracies occurring with respect to

the inconsistent or incomplete data—for instance, our

approach essentially tries to “guess” the country for

external afﬁliations by using text parsing and heuris-

tics, however, the results are not always perfect. We

acknowledge this issue within the tool documentation

and instruct the users to refer to the underlying publi-

cation records, for instance.

With respect to the interactive visual interface, we

acknowledge the potential issues with the network

view, such as the exact layout reproducibility con-

cerns after a series of interactions, or the classic trade-

offs between the aesthetic criteria (Purchase, 2002)

such as edge crossings vs. edge length. Given the dy-

namic nature of exploration and the dense relation-

ships between the network nodes, an ideal solution

is most likely not achievable here, but further efforts

should still be made, accompanied with user stud-

ies (Lam et al., 2012; Purchase, 2012) to evaluate

and compare the resulting design alternatives, for in-

stance. Future work for the interactive visual interface

also includes further support for temporal analyses of

data subsets. Finally, we see opportunities for apply-

ing text mining and visual text analytics (Kucher and

Kerren, 2015; Alharbi and Laramee, 2019) for publi-

cation abstracts or full texts as part of our future work.

7 CONCLUSIONS

In this paper, we present a project in supporting uni-

versity research and administration via interactive vi-

sual exploration of bibliographic data. We describe

the stakeholders and design requirements for this

work, including the constraints related to the use of

the existing publication data available from the biblio-

graphic data management platform DiVA established

at our university. The implemented tool addresses

the respective data peculiarities and provides multi-

ple perspectives and interactions for exploring and

investigating the publication data, the respective au-

thors/editors, their afﬁliations, and a variety of addi-

tional data facets. The tool has been used by the uni-

versity library staff to identify data quality issues with

the publication data, and by the university administra-

tion to gain understanding of the research output of

particular local research groups and environments as

well as their collaborations outside of the university,

among other application scenarios.

ACKNOWLEDGEMENTS

This research was partially supported by Linnaeus

University. The authors would like to thank Gunn

Jensen, Ida Ahlstr

om, Solbritt Andersson, and Ted

Gunnarsson for their input and support in this project.

The authors would also like to thank yWorks GmbH

for the academic license support for yFiles for HTML.

IVAPP 2023 - 14th International Conference on Information Visualization Theory and Applications

254

REFERENCES

Alharbi, M. and Laramee, R. S. (2019). SoS TextVis:

An extended survey of surveys on text visualization.

Computers, 8(1).

orner, K., Maltese, A., Balliet, R. N., and Heimlich, J.

(2016). Investigating aspects of data visualization

literacy using 20 information visualizations and 273

science museum visitors. Information Visualization,

15(3):198–213.

Burd, R., Espy, K. A., Hossain, M. I., Kobourov, S., Mer-

chant, N., and Purchase, H. (2018). GRAM: Global

research activity map. In Proceedings of the 2018

International Conference on Advanced Visual Inter-

faces, AVI ’18. ACM.

Chatzimparmpas, A., Martins, R. M., Jusuﬁ, I., Kucher, K.,

Rossi, F., and Kerren, A. (2020). The state of the art

in enhancing trust in machine learning models with

the use of visualizations. Computer Graphics Forum,

39(3):713–756.

Chen, C. (2006). CiteSpace II: Detecting and visualizing

emerging trends and transient patterns in scientiﬁc lit-

erature. Journal of the American Society for Informa-

tion Science and Technology, 57(3):359–377.

Dattolo, A. and Corbatto, M. (2022). Assisting researchers

in bibliographic tasks: A new usable, real-time tool

for analyzing bibliographies. Journal of the As-

sociation for Information Science and Technology,

73(6):757–776.

Federico, P., Heimerl, F., Koch, S., and Miksch, S. (2017).

A survey on visual approaches for analyzing scientiﬁc

literature and patents. IEEE Transactions on Visual-

ization and Computer Graphics, 23(9):2179–2198.

Felix, C., Franconeri, S., and Bertini, E. (2018). Tak-

ing word clouds apart: An empirical investigation

of the design space for keyword summaries. IEEE

Transactions on Visualization and Computer Graph-

ics, 24(1):657–666.

Fisher, D. (2005). Using egocentric networks to understand

communication. IEEE Internet Computing, 9(5):20–

28.

Heimerl, F., Han, Q., Koch, S., and Ertl, T. (2016). Ci-

teRivers: Visual analytics of citation patterns. IEEE

Transactions on Visualization and Computer Graph-

ics, 22(1):190–199.

Jo, S., Park, B., Lee, S., and Kim, J. (2021). OLGAVis: On-

line graph analysis and visualization for bibliographic

information network. Applied Sciences, 11(9).

Kucher, K. and Kerren, A. (2015). Text visualization tech-

niques: Taxonomy, visual survey, and community in-

sights. In Proceedings of the IEEE Paciﬁc Visual-

ization Symposium, PaciﬁcVis ’15, pages 117–121.

IEEE.

Lam, H., Bertini, E., Isenberg, P., Plaisant, C., and Carpen-

dale, S. (2012). Empirical studies in information vi-

sualization: Seven scenarios. IEEE Transactions on

Visualization and Computer Graphics, 18(9):1520–

1536.

Liu, J., Tang, T., Wang, W., Xu, B., Kong, X., and Xia, F.

(2018). A survey of scholarly data visualization. IEEE

Access, 6:19205–19221.

uller, E., Klosa, U., Andersson, S., and Hansson, P.

(2003). The DiVA project: Development of an elec-

tronic publishing system. D-Lib Magazine, 9(11).

Purchase, H. C. (2002). Metrics for graph drawing aes-

thetics. Journal of Visual Languages & Computing,

13(5):501–516.

Purchase, H. C. (2012). Experimental Human-Computer

Interaction: A Practical Guide with Visual Examples.

Cambridge University Press.

Roberts, J. C. (2007). State of the art: Coordinated & mul-

tiple views in exploratory visualization. In Proceed-

ings of the Fifth International Conference on Coor-

dinated and Multiple Views in Exploratory Visualiza-

tion, CMV 2007, pages 61–71. IEEE.

Rosenthal, P., M

uller, N., and Bolte, F. (2019). Visual an-

alytics of bibliographical data for strategic decision

support of university leaders: A design study. In Pro-

ceedings of the 14th International Joint Conference

on Computer Vision, Imaging and Computer Graphics

Theory and Applications—Volume 3: IVAPP, pages

297–305. SciTePress.

Russell, D. M. (2016). Simple is good: Observations of

visualization use amongst the Big Data digerati. In

Proceedings of the International Working Conference

on Advanced Visual Interfaces, AVI ’16, pages 7–12.

ACM.

Salaba, A. and Mer

cun, T. (2020). Visualizations of biblio-

graphic information: A user experience study. Journal

of Librarianship and Information Science, 52(1):271–

287.

Salinas, M., Giorgi, D., Ponchio, F., and Cignoni, P. (2020).

ReviewerNet: A visualization platform for the selec-

tion of academic reviewers. Computers & Graphics,

89:77–87.

Sedlmair, M., Meyer, M., and Munzner, T. (2012). Design

study methodology: Reﬂections from the trenches and

the stacks. IEEE Transactions on Visualization and

Computer Graphics, 18(12):2431–2440.

Small, H. (1999). Visualizing science by citation mapping.

Journal of the American Society for Information Sci-

ence, 50(9):799–813.

Small, H. (2006). Tracking and predicting growth areas in

science. Scientometrics, 68(3):595–610.

Stadler, K. (2017). The country converter coco — A Python

package for converting country names between differ-

ent classiﬁcation schemes. Journal of Open Source

Software, 2(16):332.

egas, F. B. and Wattenberg, M. (2008). Tag clouds and

the case for vernacular visualization. Interactions,

15(4):49–52.

Wiese, R., Eiglsperger, M., and Kaufmann, M. (2004).

yFiles—Visualization and automatic layout of

graphs. In Graph Drawing Software, pages 173–191.

Springer.

Zeng, W., Dong, A., Chen, X., and Cheng, Z.-l. (2021).

VIStory: Interactive storyboard for exploring visual

information in scientiﬁc publications. Journal of Vi-

sualization, 24(1):69–84.

Supporting University Research and Administration via Interactive Visual Exploration of Bibliographic Data

255