![](bg8.png)
5.4 Implementation and Experimental
Study
The given scenario, as well as a preliminary version
of CAKMAS Platform, was implemented in order to
foresee the improvements of using the proposed ap-
proaches for context capturing. As a continuation of
the previous CAKMAS implementation, it was im-
plemented in Java, based on the framework Semanti-
Core, using Jena to handle ontologies.
The scenario was run with and without the mech-
anism introduced in CAKMAS Platform. In the case
without the SBA, BobAgent gets informed when the
OA disseminates the new context in the environment,
while it persists it in the knowledge base. Each case
was run 40 times to take averages. The detailed ex-
perimental results and analysis are out of scope of
this paper given the paper length limitation. However,
just to give a glimpse of the results, the average time
BobAgent took to be aware of Alice’s action without
the SBA was 636,03 milliseconds, whereas the same
with SBA was 58,29 milliseconds. The improvement
in the time to be aware of context is significant for the
scenario, and we can deduce that it must reflect in a
real environment.
6 CONCLUSIONS AND FUTURE
WORKS
Proposing solutions for context aware virtual environ-
ments is justified by the increasing demand for tools
to support collaboration among people geographically
distributed. In such cases, the lack of a common en-
vironment in which people can act, interact and be-
come aware of the situation demands some kind of
software system to improve coordination and aware-
ness. CAKMAS Platform is proposed as an exten-
sion of CAKMAS architecture taking into account the
concept of environments, agent platforms and context
capturing. As presented in Monte-Alto et al. (2013),
previous work has focused on reasoning, persistence
and dissemination of context information, but lacked
proper ways of capturing it.
The proposed solution was analysed and validated
by means of a case study and an experimental study
using a prototype. Such empirical studies allowed
us to figure out in more detail the functionality and
effectiveness of CAKMAS Platform. However, fur-
ther validation including a more robust implemen-
tation and experimental study is necessary to give
strength to our proposal, which will be covered in fu-
ture works.
REFERENCES
Baldauf, M., Dustdar, S., and Rosenberg, F. (2007). A sur-
vey on context-aware systems. International Journal
of Ad Hoc and Ubiquitous Computing, 2(4):263–277.
Benford, S., Greenhalgh, C., Rodden, T., and Pycock, J.
(2001). Collaborative virtual environments. Commu-
nications of the ACM, 44(7):79 85.
Blois, M., Escobar, M., and Choren, R. (2007). Using
agents and ontologies for application development on
the semantic web. J. Braz. Comp. Soc., 13(2):35–44.
Chen, H. (2004). An Intelligent Broker Architecture for Per-
vasive Context-Aware Systems. PhD thesis, University
of Maryland, Baltimore County.
Churchill, E. F. and Snowdon, D. (1998). Collaborative vir-
tual environments: an introductory review of issues
and systems. Virtual Reality, 3(1):3–15.
FIPA (2000). FIPA agent management
specification. Specification 23, FIPA.
http://www.fipa.org/specs/fipa00023/SC00023K.pdf.
Khedr, M. and Karmouch, A. (2005). Acai: agent-based
context-aware infrastructure for spontaneous appli-
cations. J. Network and Computer Applications,
28(1):19–44.
Monte-Alto, H. H. L. C., Bias
˜
ao, A. B., Teixeira, L. O., and
Huzita, E. H. M. (2013). Multi-agent and context-
aware solutions for a global software development en-
vironment. International Journal of Artificial Intelli-
gence, 11(A13):115–129.
Russell, S. J. and Norvig, P. (2003). Artificial Intelligence:
A Modern Approach. Pearson Education Inc, Upper
Saddle River, NJ, 2 edition.
Tapia, D. I., Rodr
´
ıguez, S., Bajo, J., and Corchado, J. M.
(2008). Fusion@, a soa-based multi-agent architec-
ture. In DCAI, volume 50 of Advances in Soft Com-
puting, pages 99–107. Springer.
Viterbo, J., Mazuel, L., Charif, Y., Endler, M., Sabouret,
N., Breitman, K., Seghrouchni, A. E. F., and Briot,
J.-P. (2009). Managing distributed and heterogeneous
context for ambient intelligence. Context-Aware Self
Managing Systems, CRC Studies in Informatics Se-
ries, pages 79–128.
W3C (2001). W3c semantic web.
Wooldridge, M. (2002). Introduction to MultiAgent Sys-
tems. John Wiley and Sons.
ICEIS2014-16thInternationalConferenceonEnterpriseInformationSystems
612