Thomas: Practical Applications of Agents and Multiagent Systems
513
This special session has been supported by the THOMAS research project
(TIN2006-14630-C03-03), which aim is to advance and contribute methods,
techniques and tools for open multiagent systems, principally in the aspects related to
organisational structures. THOMAS is a coordinated project in which the University
of Salamanca, the Technical University of Valencia and the University of Rey Juan
Carlos cooperate to find new solutions in the field of the multiagent systems. This
special session provides a framework to disseminate the results obtained in the project
and to exchange knowledge with other researchers in the field of the agent
technology.
2 Special Session on Practical Applications of Agents and
Multiagent Systems Details
This volume presents the papers that have been accepted for the 2009 edition. These
articles capture the most innovative results and this year’s trends: Multi-Agent
Systems (MAS) Applications: commerce, health care, industry, internet, etc.; Agent
and MAS architectures; Agent development tools; MAS middleware; Agent
languages; Engineering issues of MAS; Web services and agents; Agents and grid
computing; Real-time multi-agent systems; Agent-based social simulation; Security in
MAS; Trust and reputation in MAS; Improving user interfaces and usability with
agents; Information recovery with MAS; Knowledge management with MAS;
Software Agents in Ubiquitous Computing; Agent technologies for Ambient
Intelligence; Software Agents in Industry; Planning and scheduling in MAS; Agent
Technologies for Production Systems; Service-Oriented Computing and Agents;
Agents for E-learning and education; Mobile computation and mobile
Communications. Each paper has been reviewed by three different reviewers, from an
international committee composed of 15 members from 7 different countries, and the
members of the IWANN 2009 committee. From the 22 submissions received, 17 were
selected for full presentation at the conference.
3 Special Session Acknowledgements
We would like to thank all the contributing authors, as well as the members of the
Program Committee and the Organizing Committee for their hard and highly valuable
work. Their work has helped to contribute to the success of this special session. We
also would like to thank the IWANN 2009 for giving us the opportunity of organizing
the special session, for their help and support. Thanks for your help, the special
session on practical applications of agents and multiagent systems wouldn’t exist
without your contribution.
Acknowledgments. This work has been supported by the MEC TIN2006-14630-C03-
03 project.
J. Cabestany et al. (Eds.): IWANN 2009, Part I, LNCS 5517, pp. 221–228, 2009.
© Springer-Verlag Berlin Heidelberg 2009
Self Organized Dynamic Tree Neural Network
Juan F. De Paz, Sara Rodríguez, Javier Bajo, Juan M. Corchado,
and Vivian López
Departamento de Informática y Automática, Universidad de Salamanca
Plaza de la Merced s/n, 37008, Salamanca, España
{fcofds,srg,jbajope,corchado,vivian}@usal.es
Department of Computer Science and Automation, University of Salamanca Plaza de la
Merced s/n, 37008, Salamanca, Spain
Abstract. Cluster analysis is a technique used in a variety of fields. There are
currently various algorithms used for grouping elements that are based on
different methods including partitional, hierarchical, density studies, probabilistic,
etc. This article will present the SODTNN, which can perform clustering by
integrating hierarchical and density-based methods. The network incorporates the
behavior of self-organizing maps and does not specify the number of existing
clusters in order to create the various groups.
Keywords: Clustering, SOM, hierarchical clustering, PAM, Dendrogram.
1 Introduction
The assignment of a set of objects into clusters is a widely spread problem that has
been the object of investigation in various scientific branches including
bioinformatics [10], surveillance [15], [16], [17]. Although occasionally the number
of groups is known beforehand, clustering data requires an additional step for
identifying the existing groups. There are currently different methods for creating
clusters, most notably those based on partitioning, such as k-means [11], and PAM
[9] (Partition around medoids), which work by minimizing the error function. Other
widely accepted methods are the hierarchical methods which include dendrograms
[7], agnes [9], and Diana [9]. In addition to the hierarchical methods, there are others
that use density-based models, or probabilistic-based models such as EM [8]
(Expectation-maximization) and fanny [9].
This research presents the new Self Organized Dynamic tree neural network which
allows data to be grouped automatically, without having to specify the number of
existing clusters. The SODTNN uses algorithms to detect low density zones and
graph theory procedures in order to establish a connection between elements. This
would allow connections to be established dynamically, thus avoiding the need for the
network to expand and adjust the data surface. Additionally, the connections would
continue to adapt throughout the learning process, reducing the high density neuron
areas and separating them from the low density areas.
The SODTNN integrates techniques from hierarchical and density-based models
that allow the grouping and division of clusters according to the changes in the
228
J.F. De Paz et al.
Acknowledgements. This development has been supported by the projects
SA071A08 and SIAAD-TSI-020100-2008-307.
References
[1] Furao, S., Ogura, T., Hasegawa, O.: An enhanced self-organizing incremental neural
network for online unsupervised learning. Neural Networks 20, 893–903 (2007)
[2] Kohonen, T.: Self-organized formation of topologically correct feature maps. Biological
Cybernetics, 59–69 (1982)
[3] Fritzke, B.: A growing neural gas network learns topologies. In: Tesauro, G., Touretzky,
D.S., Leen, T.K. (eds.) Advances in Neural Information Processing Systems, vol. 7, pp.
625–632 (1995)
[4] Martinetz, T., Schulten, K.: A neural-gas network learns topologies. Artificial Neural
Networks 1, 397–402 (1991)
[5] Carpenter, G.A., Grossberg, S.: The ART of adaptive pattern recognition by a self-
organizing neural network. IEEE Trans. Computer, 77–88 (1987)
[6] Shen, F.: An algorithm for incremental unsupervised learning and topology
representation. Ph.D. thesis. Tokyo Institute of Technology (2006)
[7] Saitou, N., Nie, M.: The neighbor-joining method: A new method for reconstructing
phylogenetic trees. Mol. Biol. 4, 406–425 (1987)
[8] Xu, L.: Bayesian Ying–Yang machine, clustering and number of clusters. Pattern
Recognition Letters 18(11-13), 1167–1178 (1997)
[9] Kaufman, L., Rousseeuw, P.J.: Finding Groups in Data: An Introduction to Cluster
Analysis. Wiley, New York (1990)
[10] Corchado, J.M., De Paz, J.F., Rodríguez, S., Bajo, J.: Model of experts for decision
support in the diagnosis of leukemia patients. Artificial Intelligence in Medicine (in
Press)
[11] Hartigan, J.A., Wong, M.A.: A K-means clustering algorithm. Applied Statistics 28, 100–
108 (1979)
[12] Campos, R., Ricardo, M.: A fast algorithm for computing minimum routing cost spanning
trees 52(17), 3229–3247 (2008)
[13] Bajo, J., De Paz, J.F., De Paz, Y., Corchado, J.M.: Integrating case-based planning and
RPTW neural networks to construct an intelligent environment for health care. Expert
Systems with Applications 36(3) (2009)
[14] UC Irvine Machine Learning Repository, http://archive.ics.uci.edu/
[15] Patricio, M.A., Carbó, J., Pérez, O., García, J., Molina, J.M.: Multi-Agent Framework in
Visual Sensor Networks. EURASIP Journal on Advances in Signal Processing, special
issue on Visual Sensor Networks, 21 (2007)
[16] Carbó, J., Molina, J.M., Dávila, J.: Fuzzy Referral based Cooperation in Social Networks
of Agents. AI Communications 18(1), 1–13 (2005)
[17] García, J., Berlanga, A., Molina, J.M., Casar, J.R.: Methods for Operations Planning in
Airport Decision Support Systems. Applied Intelligence 22(3), 183–206 (2005)
[18] Pavón, J., Arroyo, M., Hassan, S., Sansores, C.: Agent-based modelling and simulation
for the analysis of social patterns. Pattern Recognition Letters 29, 1039–1048 (2008)
[19] Pavón, J., Gómez, J., Fernández, A., Valencia, J.: Development of intelligent multi-sensor
surveillance systems with agents. Robotics and Autonomous Systems 55(12), 892–903
(2007)
J. Cabestany et al. (Eds.): IWANN 2009, Part I, LNCS 5517, pp. 1256–1263, 2009.
© Springer-Verlag Berlin Heidelberg 2009
Stereo-MAS: Multi-Agent System for Image Stereo
Processing
Sara Rodríguez
1
, Juan F. De Paz
1
, Javier Bajo
2
, Dante I. Tapia
1
, and Belén Pérez
1
1
University of Salamanca
2
Pontifical University of Salamanca
{srg,fcofds,jbajope,dante,lancho}@usal.es
Abstract. This article presents a distributed agent-based architecture that can
process the visual information obtained by stereoscopic cameras. The system is
embedded within a global project whose objective is to develop an intelligent
environment for location and identification within dependent environments that
merge with other types of technologies. Vision algorithms are very costly and
take a lot of time to respond, which is highly inconvenient if we consider that
many applications can require action to be taken in real time. An agent
architecture can automate the process of analyzing images obtained by cameras,
and optimize the procedure.
Keywords: Stereoscopy, stereo cameras, artificial vision, MAS, agents,
correspondence analysis, dependent environments.
1 Introduction
One of the greatest challenges for Europe and the scientific community is to find
more effective means of providing care for the growing number of people that make
up the disabled and elderly sector. The importance of developing new and more cost
effective methods for administering medical care and assistance to this sector of the
population is underscored when we consider the current tendencies. Multi-agent
systems (MAS) and intelligent device based architectures have been examined
recently as potential medical care supervisory systems [1][7][6][3] for elderly and
dependent persons, given that they could provide continual support in the daily lives
of these individuals.
The study of artificial vision, specifically stereoscopic vision, has been the object
of considerable attention within the scientific community over the last few years.
Image processing applications are varied and include aspects such as remote
measurements, biomedical images analysis, character recognition, virtual reality
applications, and enhanced reality in collaborative systems, among others.
The main topic of our research is part of a larger, global project whose objective is to
develop a system for the care and supervision of patients in dependent environments,
providing an environment capable of automatically carrying out location, identification
and patient monitoring tasks. Such an environment would also allow medical personnel
to supervise patients and simulate situations remotely via a virtual environment. In order
to reach this objective, artificial intelligence techniques, intelligent agents and wireless
technologies are used.
Stereo-MAS: Multi-Agent System for Image Stereo Processing
1263
3. Bajo, J., de Paz, J.F., de Paz, Y., Corchado, J.M.: Integrating Case-based Planning and
RPTW Neural Networks to Construct an Intelligent Environment for Health Care. Expert
Systems with Applications, part 2 36(3), 5844–5858 (2009)
4. Brenner, W., Wittig, H., Zarnekow, R.: Intelligent software agents: Foundations and
applications. Springer, Secaucus, NJ, USA (1998)
5. Canny, J.: A computational approach to edge detection. IEEE Trans Pattern Analysis and
Machine Intelligence 8(6), 679–698 (1986)
6. Corchado, J.M., Bajo, J., Abraham, A.: GERAmI: Improving the delivery of health care.
IEEE Intelligent Systems 23(2), 19–25 (2008)
7. Corchado, J.M., Bajo, J., De Paz, Y., Tapia, D.I.: Intelligent Environment for Monitoring
Alzheimer Patients, Agent Technology for Health Care. Decision Support Systems.
Elsevier Science, Amsterdam (2006)
8. Corchado, J.M., Glez-Bedia, M., de Paz, Y., Bajo, J., de Paz, J.F.: Replanning mechanism
for deliberative agents in dynamic changing environments. Computational
Intelligence 24(2), 77–107 (2008)
9. Dhond, U.R., Aggarwal, J.K.: Structure From Stereo - A Review. IEEE Trans. on Systems,
Man. and Cybernetics 19(6) (November/December 1989)
10. López-Valles, et al.: Revista Iberoamericana de Inteligencia Artificial 9(27), 35–62
(2005), ISSN: 1137-3601
11. Marr, D., Poggio, T.: A computational theory of human stereo vision. Proc. R. Soc. Lond.,
301–328 (1979)
12. Marr, D., Hildreth, E.C.: Theory of edge detection. Proc. Roy. Soc. London B-207, 187–
217 (1980)
13. Pearson, Don: Image Processing. McGrawHill, Great Britain (1991)
14. Pecora, F., Cesta, A.: Dcop for smart homes: A case study. Computational
Intelligence 23(4), 395–419 (2007)
15. Point Grey Research Inc. (2008), http://www.ptgrey.com/
16. Pollard, S.B., Mayhew, J.E.W., Frisby, J.P.: A stereo correspondence algorithm using a
disparity gradient constraint. Perception 14, 445–470 (1985)
17. Rao, A.S., Georgeff, M.P.: BDI agents: from theory to practice. In: Proceedings of ICMAS
1995, San Francisco, CA, USA (1995)
18. Rubio De Lemus, P.: Aplicacion De Una Metodologia De Evaluacion De Sistemas De
Emparejamiento En Vision Tridimensional. Psicothema ISSN edición en papel: 0214-
9915 5(1), 135–159 (1993)
19. Sobel, I., Feldman, G.: A 3x3 Isotropic Gradient Operator for Image Processing. In:
Presentado en la conferencia Stanford Artificial Project (1968)
20. Triclops StereoVision System Manual Version 3.1. User guide and command reference.
Point Grey Research Inc. (2003)
S. Omatu et al. (Eds.): IWANN 2009, Part II, LNCS 5518, pp. 963–970, 2009.
© Springer-Verlag Berlin Heidelberg 2009
CBR System with Reinforce in the Revision Phase for the
Classification of CLL Leukemia
Juan F. De Paz, Sara Rodríguez, Javier Bajo, and Juan M. Corchado
Departamento de Informática y Automática, Universidad de Salamanca
Plaza de la Merced s/n, 37008, Salamanca, España
{fcofds, srg, jbajope, corchado}@usal.es
Abstract. Microarray technology allows measuring the expression levels of
thousands of genes providing huge quantities of data to be analyzed. This fact
makes fundamental the use of computational methods as well as new intelligent
algorithms. This paper presents a Case-based reasoning (CBR) system for
automatic classification of microarray data. The CBR system incorporates novel
algorithms for data classification and knowledge discovery. The system has
been tested in a case study and the results obtained are presented.
Keywords: Case-based Reasoning, CLL, luekemia, HG U133.
1 Introduction
The use of microarrays, and more specifically expression arrays, enables the analysis
of different sequences of oligonucleotides [1], [2]. Simply put a microarray is an array
of probes that contains genetic material with a predetermined sequence. These se-
quences are hybridized with the genetic material of patients, thus allowing the detec-
tion of genetic mutations through the analysis of the presence or absence of certain
sequences of genetic material. This work focuses on the levels of expression for the
different genes, as well as on the identification of the probes that characterize the
genes and allow the classification into groups.
The analysis of expression arrays is called expression analysis. An expression
analysis basically consists of three stages: normalization and filtering; clustering and
classification; and extraction of knowledge. These stages are carried out from the
luminescence values found in the probes. Presently, the number of probes containing
expression arrays has increased considerably to the extent that it has become neces-
sary to use new methods and techniques to analyze the information more efficiently.
There are various artificial intelligence techniques such as artificial neural networks
[4], [5], Bayesian networks [6], and fuzzy logic [7] which have been applied to mi-
croarray analysis. While these techniques can be applied at various stages of expres-
sion analysis, the knowledge obtained cannot be incorporated into successive tests
and included in subsequent analyses.
This paper presents a system based on CBR which uses past experiences to solve
new problems [8], [9]. As such, it is perfectly suited for solving the problem at hand.
In addition, CBR makes it possible to incorporate the various stages of expression
analysis into the reasoning cycle of the CBR, thus facilitating the creation of
970
J.F. De Paz et al.
[3] Affymetrix. GeneChip® Human Genome U133 Arrays,
http://www.affymetrix.com/support/technical/datasheets/
hgu133arrays_datasheet.pdf
[4] Sawa, T., Ohno-Machado, L.: A neural network based similarity index for clustering
DNA microarray data. Computers in Biology and Medicine 33(1), 1–15 (2003)
[5] Bianchia, D., Calogero, R., Tirozzi, B.: Kohonen neural networks and genetic classifica-
tion. Mathematical and Computer Modelling 45(1-2), 34–60 (2007)
[6] Baladandayuthapani, V., Ray, S., Mallick, B.K.: Bayesian Methods for DNA Microarray
Data Analysis. Handbook of Statistics 25(1), 713–742 (2005)
[7] Avogadri, R., Valentini, G.: Fuzzy ensemble clustering based on random projections for
DNA microarray data analysis. Artificial Intelligence in Medicine (in press)
[8] Kolodner, J.: Case-Based Reasoning. Morgan Kaufmann, San Francisco (1993)
[9] Riverola, F., Díaz, F., Corchado, J.M.: Gene-CBR: a case-based reasoning tool for cancer
diagnosis using microarray datasets. Computational Intelligence 22(3-4), 254–268 (2006)
[10] Rodríguez, S., De Paz, J.F., Bajo, J., Corchado, J.M.: Applying CBR Systems to Microar-
ray Data Classification. In: IWPACBB 2008. Advances in Soft Computing, vol. 49, pp.
102–111 (2008)
[11] Corchado, J.M., De Paz, J.F., Rodríguez, S., Bajo, J.: Model of Experts for Decision Sup-
port in the Diagnosis of Leukemia Patients. Artificial Intelligence in Medicine (in press)
[12] Furao, S., Ogura, T., Hasegawa, O.: An enhanced self-organizing incremental neural net-
work for online unsupervised learning. Neural Networks 20(8), 893–903 (2007)
[13] Kaufman, L., Rousseeuw, P.J.: Finding Groups in Data: An Introduction to Cluster
Analysis. Wiley, New York (1990)
[14] Borg, I., Groenen, P.: Modern multidimensional scaling theory and applications.
Springer, New York (1997)
[15] Avogadri, R., Valentini, G.: The Corresponding Author and Giorgio Valentini Fuzzy en-
semble clustering based on random projections for DNA microarray data analysis. Artifi-
cial Intelligence in Medicine (in press)
[16] Vogiatzis, D., Tsapatsoulis, N.: Active learning for microarray data. International Journal
of Approximate Reasoning 47(1), 85–96 (2008)
[17] Foon, K.A., Rai, K.L., Gale, R.P.: Chronic lymphocytic leukemia: new insights into biol-
ogy and therapy. Annals of Internal Medicine 113(7), 525–539 (1990)
[18] Chronic Lymphocytic Leukemia. The leukemia and lymphoma society (2008),
http://www.leukemia-lymphoma.org/all_page.adp?item_id=7059
[19] Jurečkováa, J., Picek, J.: Shapiro–Wilk type test of normality under nuisance regression
and scale. Computational Statistics & Data Analysis 51(10), 5184–5191 (2007)
[20] Yang, T.Y.: Efficient multi-class cancer diagnosis algorithm, using a global similarity
pattern. Computational Statistics & Data Analysis (in press)
Document Outline - img945
- iwann_Thomas
- iwann_de_paz
- iwann_WSN
- iwann_stereo
- iwann_DIAMI
- iwann-iwpacbb_de_paz
- img946
Dostları ilə paylaş: |