THE SNARE ARCHITECTURE OVERVIEW
Social Network Analysis and Reengineering Environment
Alexandre Barão and Alberto Rodrigues da Silva
Instituto Superior Técnico/INESC-ID, R.Alves Redol,9,1000-029 Lisboa, Portugal
Keywords: Social network analysis, modeling, social network reengineering, transformations, ETL.
Abstract: Social network systems identify existing relations between social entities and provide a set of automatic
inferences on these relations, promoting better interactions and collaborations between these entities.
However, we find that most of existing organizational information systems do not provide in a native way,
social network features, even though they have to manage somehow social entities. The focus on this paper
starts from this fact, and proposes the SNARE system, short for “Social Network Analysis and
Reengineering Environment”. The SNARE’s purpose is to promote social network capabilities in
information systems not designed originally with that purpose. The paper overviews the architecture of
SNARE system and discusses its applicability through different approaches.
1 INTRODUCTION
Social network analysis (SNA) is focused on
describe and analyze relations between social
entities, such as people or organizations (Wasserman
and Faust, 1994). In SNA contexts, dynamics of
groups are studied to identify relations and
interactions among their members. Through SNA
it’s possible to uncover people interaction patterns
(Freeman, 2007).
SNA presents the following benefits: (1)
Improve information sharing by analyzing network
relations; (2) Increase efficiency by identifying
group or individual performances and bringing up
new roles, or providing information to redistribute
roles among groups or individuals; (3) Support
diagnostic approaches by preventing bottlenecks in
networks, e.g. evaluating periodically the weight of
information flows; and (4) Supply measures to
evaluate the impact of network changes. These
benefits are crucial for decision-making and
consequently can promote innovation and
productivity in competitive organizations.
SNA traditional studies use much information
residing in archives that were not created expressly
for social research. Sometimes, such data provide
measures of social ties and trace relationships of
actors who are reluctant to interviews. Archival data
are often inexpensive, especially when in electronic
form if maintained over time, these archives can
support SNA studies.
The validity of archival data rests on the
correspondence between measured connections and
the conceptual ties of research interest (Carrington,
P. et al., 2005). Most social networks are domain
specific, or, on the other hand, too much generic to
hold relevant data. Thus, to solve specific problems,
it is necessary to develop integration mechanisms to
allow inferences on new social networks with
different relations and topologies.
Our project, “Social Network Analysis and
Reengineering Environment”, SNARE system has
the purpose to provide social network features in
information systems not designed originally for the
effect. Through the implementation of SNARE
Social Network Metamodel, it is possible to do real
time social analysis, i.e. analyze actors continuously
and multiple relations among these actors.
Information does not flow unchanged through a
human network. People add context, interpretation,
and meaning as they receive information and pass it
along (Cross and Parker, 2004). To minimize this
fact, we assume that are actions associated to
specific relations and SNARE provides an interface
for real time actions registration. This fact enables
automatic online social analysis between actors. We
also assume that an information flow can be a
special kind of action triggered by actors.
491
Barão A. and Rodrigues da Silva A. (2008).
THE SNARE ARCHITECTURE OVERVIEW - Social Network Analysis and Reengineering Environment.
In Proceedings of the Fourth International Conference on Web Information Systems and Technologies, pages 491-494
DOI: 10.5220/0001521004910494
Copyright
c
 SciTePress

The SNARE metamodel allow researchers to
dynamically construct real scenarios for SNA
extraction and relational knowledge discovery.
We are currently considering applying SNARE
into educational contexts. To introduce SNA
features into these information systems, it is
necessary to introduce ETL (acronym for Extract,
Transform, and Load), a process that involves
extracting data from outside sources, transforming it
to fit business needs, and ultimately loading it into
the data warehouse.
Social networks can be represented according
different notations such as graphs, matrices, or
algebraic notations (Wasserman and Faust, 1994).
The data comprising social networks tend to be
heterogeneous, multirelational, and semi-structured.
Link mining is a confluence of research in social
networks, link analysis, hypertext and Web mining,
graph mining, relational learning, and inductive
logic programming (Han and Kamber, 2006).
In this paper, we introduce the motivation and
the context for social networks. Section 2 overviews
the SNARE architecture. Section 3 discusses
application scenarios of the SNARE framework.
Finally, section 4 presents preliminary conclusions
of the investigation.
2 SNARE ARCHITECTURE
The aim of the SNARE system is to extract and
analyze social networks starting from information
systems not designed originally with SNA features.
This has several benefits, e.g. reduce survey
dependent procedures, maximize reutilization of
archival data, and provide a mechanism for ongoing
social network analysis.
2.1 Architecture Approaches
SNARE provides mechanisms to extract data sets
from other systems. SNARE is designed to extract
data from a generic system by mapping data
extraction components.
There are two approaches in SNARE
architecture corresponding to two boundaries: (1)
Transparent, to support SNA of an information
system without the need to change it; and (2)
Intrusive, to support SNA depending on full access
to a generic information system source code with a
need to change it.
As suggested in Figure 1, Transparent approach
has two components: Information System (IS) and
SNARE ETL Services. The IS denotes a generic
information target system. SNARE ETL Services is
an engine that provides ETL features, i.e. extract
data from outside sources, transform it, and load it
into the SNARE database.
On the other hand, Intrusive approach includes
the Extended Information System (IS*) which is an
IS refinement and the SNARE Services system,
which provides a set of social network features
through web services.
Besides these two architecture approaches,
SNARE Services can be executed autonomously, i.e.
through SNARE user interface applications, it is
possible to perform social network analysis tasks
without the need of external information systems
integration.
Figure 1: SNARE Architecture Approaches.
2.2 System Artefacts
SNARE is an application under development to be
web accessed over a network such as the Internet or
an intranet.
Figure 2: SNARE system packages.
As depicted in Figure 2, SNARE system
involves a set of packages. Figure 2 also emphasizes
the social network analysis context and the data
reengineering context. Each package has a specific
function, making possible to distinguish several
stereotypes in SNARE components such as: web
application, windows application, system, and
WEBIST 2008 - International Conference on Web Information Systems and Technologies
492

finally, an engine. Figure 3 shows the SNARE
system components view.
Figure 3: SNARE system components view.
2.2.1 Analysis Components
Based on social network models, SNARE Services
provides methods to ensure the definition of a set of
models to allow the representation of social
networks.
Web SNARE is a web-based application while
Desktop SNARE is the equivalent tool for a
windows-based application. Both include four
modules: Social Network Environment Settings;
Social Network Definitions; Social Network
Instantiation; and Social Network Explorer. Through
Social Network Environment Settings, users can
customize environment settings that automate data
collection for SNA. Each user has an identity.
Authorization and security policy is defined
according to a flexible role-based mechanism. The
Social Network Definitions Module is used to create
social entities, roles, relations, actions, events, i.e.
structural SNA data. The Social Network
Instantiation Module make possible to instantiate
relations, actions, events and surveys. Surveys allow
investigators to decide on relationships to measure
and on actors/objects to be approached for data and
in the absence of archival records, surveys are often
the most practical alternative (Carrington et al.,
2005). SNARE survey instantiation interface
mechanisms are useful for entering and visualizing
personal or group social network data. Finally, the
SNARE Social Network Module Explorer is the
SNARE interface for social network data analysis
requests and visualization.
SNARE Services include a library and a
database as showed in Figure 3. SNA algorithms and
model transformations are provided by SNARE
Library, i.e. a set of transformations, allowing
conversions between models and algorithms which
may analyze the social networks to retrieve relevant
information. Components interact with each other
through provided and required interfaces. SNARE
Services provide two main (classes set) of interfaces:
(1) interfaces to support web and windows-based
applications; and (2) interfaces to support external
source integration. SNARE Services has a built-in
XML parser for SNARE Schema, which is a schema
to handle social networks relational data including
relations, actions, events, social entities and roles,
through provided interfaces (see section 3 for more
details).
Developing models to represent social networks,
allows SNARE Services implementation of specific
graph algorithms which can extract valuable
information from hidden social networks.
Algorithms from Link Analysis, an area pertaining
to the more general research field of data mining
with the purpose of extracting new information on
linked structures such as graphs, can be also applied
on social networks . These include classification
algorithms, which predict the value of nodes based
on their links, and clustering algorithms, which
attempt to group nodes of a same kind and are of
special interest in finding communities on linked
structures.
2.2.2 Data Reengineering Components
On the other hand, SNARE ETL Services provide a
technical interface to Desktop ETL Tool, a desktop
application to define and control ETL actions and
has a required interface to execute SNARE Services
methods. The aim of this component is to extract
relevant social network data through ETL
mechanisms. This tool allows users to specify
transforms through a graphical user interface.
Data transformation can involve the following:
(1) Smoothing, which works to remove noise from
data; (2) Aggregation, where summary operations
are applied to the data, typically used in constructing
a data cube for analysis of the data at multiple
granularities; (3) Generalization of the data, where
low-level data are replaced by higher-level concepts
through the use of concept hierarchies; (4)
Normalization, where the attribute data are scaled so
as to fall within a small specified range; and (5)
Attribute construction, where new attributes are
constructed and added from the given set of
attributes to help the mining process (Han and
Kamber, 2006).
Transparent approach use SNARE ETL
Services. These services are encapsulated in the
SNARE ETL library component. SNARE ETL
THE SNARE ARCHITECTURE OVERVIEW - Social Network Analysis and Reengineering Environment
493

Services use the interface to external source
integration provided by SNARE Services and
SNARE metamodel is used to define this
integration.
3 APPLICATION SCENARIOS
From our experience using and developing e-
Learning based systems, we identify several system
limitations, such as: difficulty to recognize explicit
relations between learning objects, authors, students
and teachers, or between teachers and students, or
even between students interested in a common
subject. The challenge of our research is to make
explicit these relations, supported by a generic
platform (such as SNARE). In order to enhance
learning processes, it is possible to improve actor’s
relationships through social networks inferences on
systems that were not design for that purpose.
For an e-learning scenario, we are considering
Moodle platform (“Moodle”, 2007), which is, under
a constructivist perspective, an internet-based course
management system designed using pedagogical
principles, to help educators create online learning
communities. Our project scenario project uses
Moodle platform where users can freely sign-up and
create educational contents using learning objects. In
this scenario, we should apply the SNARE
Transparent approach because we do not intend to
change the Moodle source code.
To infer social networks on other systems,
another scenario would be considered. This scenario
is a Learning Management System to support school
management activities, involving different actors,
e.g. students, teachers, educators or parents. In this
scenario we will apply the SNARE Intrusive
approach.
4 CONCLUSIONS
This paper introduces the problems and motivation
behind our research work and overviews the
proposed SNARE system.
Throughout this work, we propose a set of
components to analyze social networks from real
application scenarios. The main purpose of SNARE
is to analyze social networks on systems not
previously designed for the effect.
Based on social network models, SNARE
Services provides methods to ensure the definition
of a set of models to allow the representation of
social networks.
With SNARE ETL Services, the system provides
ETL features in order to analyze databases with the
specific purpose to extract, transform and load data
to SNARE database. This fact allows the use of
specific graph algorithms which can extract valuable
information from hidden social networks.
SNARE metamodel ensures that relations,
actions and events can have multiple extreme
instances and the social network system keeps
references to all previous metaclasses.
From the research preliminary discussed in this
paper, we conclude that much work on the area of
social network analysis is still open, and that this
area has a growing potential that should be explored.
As a consequence of this project, we hope to provide
new approaches and technologies to improve the
organizational environment and, in particular, to
improve e-learning and scholar management systems
user interactions, to maximize educational success.
Finally, it would be of interest the development
of new systems, taking advantage of the proposed
SNARE application in other organizational contexts.
REFERENCES
Carrington, P.J., J. Scott, and S. Wasserman, 2005. Models
and methods in social network analysis. Structural
analysis in the social sciences ; 27. Cambridge ; New
York: Cambridge University Press. xiv, 328 p.
Cross, R.L. and A. Parker, 2004. The hidden power of
social networks : understanding how work really gets
done in organizations. Boston, Mass.: Harvard
Business School Press. xiii, 213 p.
Freeman, Linton C., 2007. What is Network Analysis.
International Network for Social Network Analysis.
Accessed in November 2007. Available at:
http://www.insna.org/INSNA/na_inf.html
Han, J. and M. Kamber, 2006. Data mining : concepts and
techniques. 2nd ed. Morgan Kaufmann series in data
management systems. Boston: Morgan Kaufmann.
xxviii, 770 p.
Moodle. 2007. http://www.moodle.org
Wasserman Stanley and Faust, Katherine, 1994. Social
Network Analysis: methods and applications. In:
Structural analysis in social the social sciences series.
Cambridge: Cambridge University Press, (1994) 2006.
ISBN 0-521-38707-8.
WEBIST 2008 - International Conference on Web Information Systems and Technologies
494