Saturday , April 20 2024

Metamodelling for Agent-Based Modelling: An Application for Posted Pricing Institutions

Ruben FUENTES-FERNÁNDEZ
GRASIA-Universidad Complutense de Madrid, Departamento de Ingenieria del Software e Inteligencia Artificial, Facultad de Informatica, c/ Profesor Jose Garcia Santesmases s/n
Madrid, 28040, Spain

Jose M. GALÁN
INSISOC-Universidad de Burgos, Escuela Politecnica Superior, c/Villadiego s/n,
Burgos, 09001, Spain

Samer HASSAN
GRASIA-Universidad Complutense de Madrid, Departamento de Ingenieria del Software e Inteligencia Artificial, Facultad de Informatica, c/ Profesor Jose Garcia Santesmases s/n
Madrid, 28040, Spain

Felix A. VILLAFANEZ
INSISOC-Universidad de Valladolid, Escuela de Ingenierias Industriales,
Paseo del Cauce 59, Valladolid, 47011, Spain

Abstract: Agent-Based Modelling is gaining wider acceptance as a paradigm for social research. However, it still presents limitations in the management of the process to generate the simulations from the initial conceptual models. This makes it difficult to reuse the knowledge from available models and to adapt it to different hypotheses. This paper proposes the use of metamodels in order to define explicitly the core concepts of a problem domain and to differentiate the aspects involved in the process. A case study for posted pricing institutions shows how to define the related metamodel and use it to address alternative situations in these auctions. The case study drives the discussion on the advantages and limitations that metamodelling can bring to social simulation.

Keywords: Accessibility, usability, heuristic evaluation, municipal web sites.

>>Full text

CITE THIS PAPER AS:

Ruben FUENTES-FERNÁNDEZ, Jose M. GALÁN, Samer HASSAN, Felix A. VILLAFANEZ, Metamodelling for Agent-Based Modelling: An Application for Posted Pricing Institutions, Studies in Informatics and Control, ISSN 1220-1766, vol. 20 (1), pp. 55-66, 2011. https://doi.org/10.24846/v20i1y201105

1. Introduction

Agent-Based Modelling (ABM) is a form of computational simulation that has become over the past years a widely used technique for research in very different disciplines such as Biology [20], Resource Management [22] or Political Science [21]. One of its key advantages lies on its core abstractions for modelling, i.e. agents and their societies. Agents are computational, intentional and social entities, which are capable of rational and complex individual behaviour. As they share these features with humans, they are expected to facilitate the specification of the human target systems, which makes them especially suitable to be used in the Social Sciences [14]. Besides, these abstractions can be refined with concepts closer to the target simulation platforms, bridging the gap between the conceptual models and the simulation ones.

This modelling approach has been easily adopted because [1, 4, 9]:

  • It leads to natural and, simultaneously, formal descriptions of the target systems that can be understood and validated by modellers and stakeholders.
  • It enables to easily model heterogeneity and bounded rationality, making possible to abandon assumptions of representative and optimizing behaviours that are non-realistic in many social contexts.
  • It facilitates to integrate an explicit representation of the environment and to model local interaction.
  • It allows analysing bottom-up and emergent behaviours.
  • It makes possible to create interdisciplinary science integrating models from different fields.

In spite of these advantages and the diffusion of the approach, the actual use of agent-based models still has a lot of room for improvement from a computational point of view [7]. The situation in which agents constitute general and reusable modelling primitives for ABM, with standardised processes to translate them to simulation models has not yet been reached. In fact, researchers in ABM have different perspectives on what agents are, which range from simple sets of key-value pairs to complex rational entities with an explicit representation of their knowledge and reasoning. Moreover, researchers do not tend to follow a clear and explicit translation of those perspectives into their formal agent models and simulation code. On the contrary, they frequently adopt ad-hoc translations whose details are not made available, as they are considered of secondary relevance for research. This state of affairs makes difficult comparing hypotheses and results, and hinders a wider acceptance of ABM [12].

In order to address these limitations, some researchers [18, 31] have proposed the use of metamodelling techniques. Metamodels define modelling languages, which specify the modelling primitives available to describe a problem. Researchers create their models instantiating these primitives, i.e. creating models that comply with the definition of the metamodel. If required, they can use extension mechanisms to modify the language in a controlled way, introducing new elements or modifying the available ones.

This approach based on metamodels has two key advantages. Firstly, metamodels can be processed by software tools. This allows, for instance, developing graphical editors for these models, and providing automated transformations that partly carry out the propagation of information from abstract formal models to actual simulations. This tool support reduces the probability of making unintended mistakes when modelling, as tools make some basic checking of the information. Moreover, this support provides the basis for comparison (and thus replication) among works, as refinement and implementation information is available for examination. Secondly, metamodels make explicit the primitives required to model certain classes of problems under a set of given assumptions. This knowledge crystallizes the experience of a community in a domain, facilitates learning to novices, and encourages discussion and reflection on different approaches.

The main obstacle to apply this approach is to define suitable metamodels and the correspondences between them. There is an inherent difficulty in capturing the knowledge to describe complete domains of problems with a formal definition. A metamodel must be rich enough to capture all the variability of a domain, but also to constrain modellers to produce correct models that can be translated to simulation code in a semi-automated way. There must also be clear correspondences between the metamodels of the different languages involved in a problem (e.g. domain, software-modelling and programming languages). This is not only an issue of making semi-automated translations, but making those proper according to the semantics of the involved elements. Some semantic alignment of the concepts in the different languages facilitates this process.

Our work addresses this problem with a framework for metamodelling in ABM. It includes intermediate languages between the abstract formal models and the simulation code, guidelines to define their metamodels and the correspondences between the different involved languages, and software tools that support these tasks. Its modelling process is conceived as a successive refinement of models in different languages, supported by transformations and tailored tools, until generating the simulation. In order to illustrate this approach, this paper discusses the formalisation of the well-known problem of posted pricing institutions.

The rest of the paper is organised as follows. Section 2 provides a brief introduction on ABM. Section 3 considers how our approach use metamodels in ABM and Section 4 applies it to model auctions. Finally, Section 5 discusses the implications of the process together with some concluding remarks.

REFERENCES

  1. ABASCAL, J., M. ARRUE, I. FAJARDO, N. GARAY, J. TOMAS, The Use of Guidelines to Automatically Verify Web Accessibility, Universal Access in the Information Society, vol. 3(1), 2004, pp. 71-79.
  2. ALECU, S., Accessibility Evaluation of a Web Application for Visually Impaired People, Revista Româna de Interactiune Om-Calculator 1 (Special Issue RoCHI 2008), 2008, pp. 15-18.
  3. ANPH (2010) Statistical Bulletin Q1. Autoritatea Naţională pentru Persoanele cu Handicap. Available at: http://www.anph.ro/eng/news.php?ida=15
  4. BIRSH, J. R., Research and Reading Disability. Multisensory Teaching of Basic Language Skills, Baltimore, Maryland, Paul H. Brookes Publishing, 2005, p. 8.
  5. CZEPITA, D., E. LODYGOWSKA, Role of the Organ of Vision in the Course of Developmental Dyslexia. Klin Oczna, vol. 108(1-3), 2006, pp. 110-3.
  6. e-Europe 2002 Action Plan, prepared by the Council and the European Commission for the Feira European Council, 19-20 June 2000, published on June 14, 2000.
  7. e-Europe 2005 Action Plan, presented in view of the Sevilla European Council, 21/22 June 2002, published on May 28, 2002.
  8. FAGA, J. C., B. FAGAN, An Accessibility Study of State Legislative Web Sites, Government Information Quarterly, vol. 21, 2004, pp. 65-85.
  9. HACKETT, S., B. PARMANTO, Homepage Not Enough when Evaluating Web Site Accessibility, Internet Research, vol. 19(1), Emerald. 2008, pp. 78-87.
  10. KANE, S., J. SHULMAN, R. LADNER, T., SHOCKELY, A Web Accessibility Report Card for Top International University Web Sites, Proceedings of W4A2007, ACM, 2007, pp. 148-156.
  11. KUZMA, J., Accessibility Design Issues with UK E-Government Sites, Government Information Quarterly, vol. 27, 2010, pp. 141-146.
  12. ISAILA, N., I. SMEUREANU, The Accessibility of Information in Computer Assisted Learning Process for Persons with Disabilities, Proceedings of ECC’2010, 2010, pp. 139-142.
  13. ISO/IEC FCD 25010:2010. Software Engineering – Software Product Quality Requirements and Evaluation (SQuaRE) – System and software quality models, JTC 1/SC 7 N4522, 2007
  14. IVAN, I., B. VINTILĂ, C. CIUREA, M. DOINEA, The Modern Development Cycle of Citizen Oriented Applications, Studies in Informatics and Control, vol. 18(3), 2009, pp. 263-270.
  15. LAZAR, J., A. DUDLEY-SPONAUGLE, K.-D. GREENIDGE, Improving Web Accessibility: A Study of Webmaster Perceptions, Computers in Human Behaviour, vol. 20, 2004, pp. 269-288.
  16. LESNEANU, I., D. D. IORDACHE, Testing with Visually Impaired Users of a Local Public Administration Web Site, Revista Româna de Interactiune Om-Calculator, vol. 3 (Special Issue RoCHI 2010), 2010, pp. 15-18.
  17. LEUTHOLD, S., J. BARGAS-AVILA, K. OPWIS, Beyond Web Content Accessibility Guidelines: Design of Enhanced Text User Interfaces for Blind Internet Users, International Journal for Human-Computer Studies, vol. 66, 2008, pp. 257-270.
  18. McCANDLISS, B. D., NOBLE K. G., The Development Of Reading Impairment: A Cognitive Neuroscience Model, Mental Retardation and Developmental Disabilities Research Reviews, vol. 9(3), 2003, pp. 196-204.
  19. MeAC, Measuring progress of eAccessibility in Europe. Available at: http://ec.europa.eu/information_society/activities/einclusion/library/studies/meac_study/index_en.htm, 2007.
  20. OLSEN, M. G., How Accessible is the Public European Web, http://www.mortengoodwin.net/publicationfiles/how_accessible_is_the_european_web.pdf, 2008.
  21. OSICEANU, M. E., GHIOC, S. The Role and the Importance of Adaptated Information Technology (AIT) in the Process of Social Integration of Persons with Disabilities, Revista Româna de Interactiune Om-Calculator, vol. 1(2), 2008, pp. 133-144.
  22. PĂDURE, M., The Accessibility of Elearning Platforms for the Visually Impaired Students, Revista Româna de Interactiune Om-Calculator, vol. 2 (Special Issue RoCHI 2009), 2009, pp. 25-30.
  23. Riga Ministerial Declaration (2006), Riga, Latvia. Available at: http://ec.europa.eu/information_society/events/ict_riga_2006/doc/declaration_riga.pdf
  24. SUDUC, A. M., M. BIZOI, F. G., FILIP, User Awareness about Information System Usability, Studies in Informatics and Control, vol. 19(2), 2010, pp. 145-152.
  25. TAKAGI, H., S. KAWANAKA, M. KOBAYASHI, D. SATO, C. ASAKAWA, Collaborative Web Accessibility Improvement: Challenges and Possibilities, Proceedings of ACM SIACCESS, ACM, 2009, pp. 195-202.
  26. Total Validator. Available at: http://www.totalvalidator.com/
  27. US Section 508 (1998). Available at: http://www.section508.gov/
  28. WAI (1997) Web Accessibility Initiative, W3C. Available at: http://www.w3.org/WAI/
  29. WCAG1 (1999) Web Content Accessibility Guidelines 1.0, W3C, 1999. Available at: http://www.w3.org/TR/WCAG10/
  30. WCAG2 (2008) Web Content Accessibility Guidelines 2.0, W3C, 2008. Available at: http://www.w3.org/TR/WCAG20/
Designed by | ICI Bucharest
© Copyright 2024, All Rights Reserved