Visualizing library data interactively: two demonstrations using R language
| DOI | https://doi.org/10.1108/LHTN-01-2018-0003 |
| Pages | 14-17 |
| Published date | 02 July 2018 |
| Date | 02 July 2018 |
| Author | Zhehan Jiang,Richard Carter |
| Subject Matter | Library & information science,Librarianship/library management,Library technology,Library & information services |
Visualizing library data interactively: two
demonstrations using R language
Zhehan Jiang and Richard Carter
Introduction
In the era of big data, the amount of
quantifiable information continues to grow
by leaps and bounds. Currently, the
volume, velocity and variety of data are
vast (Ahmed and Ameen, 2017), and
therefore, to make highly informed
decisions with agility, an organization
needs to be able to access and interpret
data in real-time. This idea is particularly
important for libraries that host records
about collections, assessment statistics and
budget figures, as well as log information
on both the internet and intranet.
Carefully monitoring and evaluating
data can help libraries better serve their
patrons and communities in accordance
with ever-changing needs. In addition,
scientifically using data for library
development provides support for
departmental and institutional accreditation.
Among all data handling approaches, data
visualization offers a straightforward yet
valuable way to discover and interpret
information hidden in the data. Unlike
calendar models that require highly specific
knowledge to interpret the results, data
visualization is more laymen-friendly and
therefore more approachable, as it does not
require advanced prerequisite skills to
comprehend. In fact, human brains are
programmed to receive signals yielded by
graphics and plots that include color
intensity, hue distribution, shape, size, line
type, enclosure and orientation (Few, 2012,
pp. 67-71). Therefore, for most audiences,
perceiving visual information transfer is
easier and more efficient than that of written
or verbal format (Dur, 2014).
Although data visualization has been
implemented in library environments for
some time, current demand for visualizing
data is unprecedentedly high because of
the greater availability and accessibility of
both data and computing facility in
libraries. This trend makes traditional
techniques, such as Excel and PowerPoint,
incapable to meet the challenges, as they
are not designed for visualizing large-scale
and/or multi-dimensional data. Modern
data visualization techniques are based
upon software applications with advanced
features such as:
database query;
cloud and Web-browser-based
connectivity;
drag and drop interfaces;
graphic style variability;
powerful analytic engines; and
compatibility with other frameworks
and software applications.
Librarians and researchers have
recognized the advantages of
implementing modern data visualization
techniques to library data. For example,
Lowery (2011) describes visualizing an
atlas collection via an online tool named
Many Eyes; Wright adopts Piktochart to
create infographics for medical libraries
information, and Murphy (2015)
delineates how Tableau was implemented
to support academic library assessment.
Further examples include S. and Naik’s
(2017) demonstrations using Viewshare to
produce graphics for E-journal data.
Details about different graphics, designs
and visualization tool comparisons can be
found in Archambault et al. (2015).
R for data visualization
Despite a large number of visualization
options that are both available and
attainable, R (R Development Core Team,
2018) is positioned to be the most flexible
tool among other software applications of
its kind. In fact, instead of being specific
for visualizing data, R is a programming
language and a software system.
With the boost of analytic needs, R is
becomingthelinguafrancafordata-
related tasks. This open source
environment enriches users’ solutions with
a great compatibility of different computer
systems, platforms, software applications
and Web frameworks. For instance, R can
be used to scrape website data, recognize
images and voices, execute application
programming interfaces (APIs), conduct
forecasting analysis and missing data
imputation, build websites, construct
algorithmic estimators and provide both
static and interactive visualization. In
addition, thanks to the support from the
rapidly expanding R community, users are
allowed to implement the most up-to-date
yet valuable functions and techniques for
free while sharing information and
problem-solving. In particular, there are
outstanding packages in R providing great
solutions to visualization tasks across the
entire data processing span (i.e. data
extraction to graphics production).
Unfortunately, to date, few librarians
have routinely used R to accomplish tasks
because of its steep learning curve; that
said, there is a trade-off between gaining
highly customized R functionalities and
practicing programming logic and skills.
Nevertheless, according to the Federal Big
Data Research and Development Strategic
Plan (Big Data Senior Steering Group,
2016), the modern roles, which librarians
will play for meeting the fast demand for
analytical talent and capacity across all
sectors of the national workforce, do
require programming knowledge.
Therefore, it is a great investment for
librarians to learn and adopt modern
programming languages such as R to
handle library data. In the following
paragraphs, two examples are described so
that readers can gain a sense of the value of
modern visualizations as well as the
flexibility of R.
Modern visualization examples
The first example is a map showing the
use of the content from the University of
The authors would like to thank the
University of Alabama Libraries for all of
the support.
14 LIBRARY HITECH NEWS Number 5 2018, pp. 14-17, V
CEmerald Publishing Limited, 0741-9058, DOI 10.1108/LHTN-01-2018-0003
To continue reading
Request your trial