By classifying business intelligence appropriately, we allow ourselves to spot opportunities for investment and exploitation, increasing our ability to turn the data and insight we collect into profit. Business intelligence and its research can be
divided into a taxonomy. This paper breaks that down. Even without data, are there areas that may contain similar opportunities?
1. Introduction and Motivation
1.5. Survey Scope
Due
to the multidisciplinary nature of this state-of-the-art report, we
require a well-defined scope to ensure that the most relevant research
is included. The scope encompasses academic research that has emphasis
placed on state-of-the-art visualisation for the purposes of business
data exploration. The quality of the visual design must be regarded as
state-of-the-art – and the motivation behind the research must focus on
business data.
Many
conferences and journals publish articles which include visualisation
of business data but do not focus on novel visual design aspects of the
research. Papers published in visualisation journals are a valuable
source for our literature scope. Non-visualisation journal and
conference papers are only in scope if the focus of the research is
visualisation-oriented with the ultimate goal of better understanding
the business data for informing the decision-making process.
The
primary source of in-scope research comes from the conferences and
journals that make visualisation the subject of their publishings. The
conference proceedings of IEEE VIS and EuroVis, or the IEEE TVCG and CGF
journal contain a wealth of publications that focus their attention on
business data. These papers are considered the primary driving force
behind the evolution of the field. See Table 3.
1.5.1. Out of Scope
Publication venues such as
the conference "Software Engineering and Service Science" (ICSESS) may
publish papers that include visualisation of business data but do not
place emphasis on novel visual design or value. For example, in a paper
entitled "Visualizations-based
Analysis of Telco Data for Business Intelligence", Ashraf and
Khan designed imagery to represent telecommunications data. However,
these images come in the form of a pie chart and a radar chart where
most of the analysis is performed through a
numerical calculation. We do not include papers like this in the
scope due to the limited visual component of the research.
Papers
we consider to be borderline might have good potential in the field of
business visualisation and often mention this as a valid application,
however, they do not focus on the business aspect. For example, Wu et
al. created an opinion-based visual design from social media data which
shows valuable public opinion on products but also on non-business based
events such as WWII and political scandals. Because of the tenuous
connection, we do not include this within scope.
The
topic of social media falls outside of the scope of this survey.
Including it would make the survey too large. We refer readers to Wanner
et al. for an existing survey on this topic.
1.5.2. In Scope
It is possible for papers in
cross-disciplinary journals to be within scope, only if the
visualisations used are state-of-the-art or add additional value to the
visualisation literature. For example, the journal "Expert Systems with
Applications"
published Hanafizadeh and Mirzazadeh's "Visualizing market
segmentation using self-organizing maps and Fuzzy Delphi method-ADSL
market of a telecommunication company", where advanced visual methods
are used with emphasis placed on the contribution
of the visualisation within a business context. A further
example of an in-scope paper comes from the proceedings of the
"International Conference on Web Intelligence and Intelligent Agent
Technology" in a paper by Ziegler et al. entitled "Mining
and Exploring Unstructured Customer Feedback Data Using Language
Models and Treemap Visualizations" where customer feedback data are
structured in a specialised treemap. This falls within scope due to the
focus on the customised treemap design
and the novel features implemented in the software.
The main body of publications in this survey was obtained from the major visualisation publication venues. However, the "business" component of each paper is more complicated to define. To clarify this aspect of the scope, we impose a heuristic that a business related subject has to be mentioned in the title or abstract of the paper. A case study alone is not enough to fall within scope. For an overview of the business-related subjects, please refer to the literature search methodology in Section 1.4.