Emilio Serrano

Testing and Debugging of MAS Interactions with INGENIAS

Testing and debugging activities are getting more relevance in multi-agent systems (MAS) as agents become part of real applications. Both activities are related, since failures to be debugged are frequently detected during the execution of tests. The support for these activities is not yet as complete as other activities of MAS development. However, agent oriented software engineering methodologies are incorporating new testing and debugging features. In this direction, the paper introduces advances made in the INGENIAS agent development framework towards a complete coverage of testing and debugging activities. The advances are compared with respect to a categorisation of related works in the agent literature. This categorisation will be useful for evaluating and planning issues for improvement in the context of INGENIAS.

Intelligent data analysis applied to debug complex software systems

The emergent behavior of complex systems, which arises from the interaction of multiple entities, can be difficult to validate, especially when the number of entities or their relationships grows. This validation requires understanding of what happens inside the system. In the case of multi-agent systems, which are complex systems as well, this understanding requires analyzing and interpreting execution traces containing agent specific information, deducing how the entities relate to each other, guessing which acquaintances are being built, and how the total amount of data can be interpreted. The paper introduces some techniques which have been applied in developments made with an agent oriented methodology, INGENIAS, which provides a framework for modeling complex agent oriented systems. These techniques can be regarded as intelligent data analysis techniques, all of which are oriented towards providing simplified representations of the system. These techniques range from raw data visualization to clustering and extraction of association rules.

Validating ambient intelligence based ubiquitous computing systems by means of artificial societies

This paper introduces a new methodology based on the use of Multi-Agent Based Simulations (MABS) for testing and validation of Ambient Intelligence based Ubiquitous Computing (UbiCom) systems. An ambient intelligence based UbiCom is a pervasive system in which services have some intelligence in order to smoothly interact with users immersed in the environment. The motivation for this methodology is its application in UbiCom large-scale systems where large numbers of users are involved and in applications which deal with dangerous environments. In these cases, real tests are impractical and an artificial society is required. MABS allows building cheap and quick prototypes which can describe UbiCom systems. Analyzing these prototypes, if they are sufficiently descriptive, allows requisites violations in functionality of real UbiCom system designs to be discovered. MABSs and particularly the most descriptive ones can present very complex behaviors. Therefore, the MABS analysis obtained with the presented methodology is not trivial. Consequently, this paper also proposes two techniques for the analysis of general complex MABSs: forensic analysis and the use of simpler simulations. Moreover, the methodology proposes to inject elements of the actual UbiCom system in the simulated world to increase the confidence of the validation process. The proposal is illustrated with a detailed case study that considers a building on our campus and an AmI service for evacuation in case of fire.

Debugging complex software systems by means of pathfinder networks

This paper introduces a new methodology based on the use of Pathfinder networks (PFNETs) for the debugging of multi-agent systems (MASs). This methodology is specifically designed to develop a forensic analysis (i.e. a debugging process performed on previously recorded data of the MAS run) of MASs showing complex tissues of relationships between agents (i.e. a high complexity in their social level). Like previous works in the field of forensic analysis of MASs, our approach is performed by considering displays of the system activity which aim to be understandable by human beings. These displays allow us to understand the social behavior of the system, discover emergent behaviors, and debug possible undesirable behaviors. However, it is well known that the visualization of information in a humanly comprehensible way becomes a complex task when large amounts of information have to be represented, as is the case of the social behavior of large-scale MASs. Our methodology tackles this problem through the use of PFNETs, which are considered to reduce the data complexity in order to obtain simple representations that show only the most important global interactions in the system. In addition, the proposed methodology is customizable thanks to the use of two thresholds allowing the user to define the desired specificity level in the display. The proposal is illustrated with a detailed case study considering a complex customer-seller MAS.

An Approach for Representing Sensor Data to Validate Alerts in Ambient Assisted Living

The mainstream of research in Ambient Assisted Living (AAL) is devoted to developing intelligent systems for processing the data collected through artificial sensing. Besides, there are other elements that must be considered to foster the adoption of AAL solutions in real environments. In this paper we focus on the problem of designing interfaces among caregivers and AAL systems. We present an alert management tool that supports carers in their task of validating alarms raised by the system. It generates text-based explanations—obtained through an argumentation process—of the causes leading to alarm activation along with graphical sensor information and 3D models, thus offering complementary types of information. Moreover, a guideline to use the tool when validating alerts is also provided. Finally, the functionality of the proposed tool is demonstrated through two real cases of alert.

An approach to debug interactions in multi-agent system software tests

One of the most recurrent approaches for testing and debugging multi-agent systems is the use of displays which show recorded interactions among agents. These displays are studied in order to discover faults in the software. Three main shortcomings are present in this approach: (1) how to capture the interactions in distributed multi-agent systems is not usually explained; (2) a total order among the events is considered and this is not accurate in a distributed system; (3) an excess of information is displayed to developers without the possibility of obtaining a summary. This paper offers a solid infrastructure to capture, order, display and summarize messages exchanged in multi-agent systems. To deal with (1), a generic registration layer is offered by using aspect oriented programming. Vector clocks are employed to order the distributed events with the aim of solving (2). These clocks are combined with graph theory to obtain simplified representations of the interactions. Finally, abstract graphs are presented as a mechanism to summarize interactions to cover (3). Several case studies demonstrate the utility of the approach presented here and an open source implementation is provided along with the paper.

Infrastructure for Forensic Analysis of Multi-Agent Systems

The contribution of this paper is an intent to state the basis for forensic analysis of multi-agent system (MAS) runs. It proposes a general approach for open source agents platforms. It consists on techniques to store, order and represent messages based on conventional observation of the events in a distributed system, particularized for the case of MAS in which agents can be distributed across a number of machines or even be mobile.

Towards a Holistic Framework for the Evaluation of Emergency Plans in Indoor Environments

One of the most promising fields for ambient intelligence is the implementation of intelligent emergency plans. Because the use of drills and living labs cannot reproduce social behaviors, such as panic attacks, that strongly affect these plans, the use of agent-based social simulation provides an approach to evaluate these plans more thoroughly. (1) The hypothesis presented in this paper is that there has been little interest in describing the key modules that these simulators must include, such as formally represented knowledge and a realistic simulated sensor model, and especially in providing researchers with tools to reuse, extend and interconnect modules from different works. This lack of interest hinders researchers from achieving a holistic framework for evaluating emergency plans and forces them to reconsider and to implement the same components from scratch over and over. In addition to supporting this hypothesis by considering over 150 simulators, this paper: (2) defines the main modules identified and proposes the use of semantic web technologies as a cornerstone for the aforementioned holistic framework; (3) provides a basic methodology to achieve the framework; (4) identifies the main challenges; and (5) presents an open and free software tool to hint at the potential of such a holistic view of emergency plan evaluation in indoor environments.

Strategies for avoiding preference profiling in agent-based e-commerce environments

Agent-based electronic commerce is known to offer many advantages to users. However, very few studies have been devoted to deal with privacy issues in this domain. Privacy is of great concern and preserving users’ privacy plays a crucial role to promote their trust in agent-based technologies. In this paper, we focus on preference profiling, which is a well-known threat to users’ privacy. Specifically, we review strategies for customers’ agents to prevent seller agents from obtaining accurate preference profiles of the former group by using data mining techniques. We experimentally show the efficacy of each of these strategies and discuss their suitability in different situations. Our experimental results show that customers can improve their privacy notably with these strategies.

A Survey of Twitter Rumor Spreading Simulations

Viral marketing, marketing techniques that use pre-existing social networks, has experienced a significant encouragement in the last years. In this scope, Twitter is the most studied social network in viral marketing and the rumor spread is a widely researched problem. This paper contributes with a survey of research works which study rumor diffusion in Twitter. Moreover, the most useful aspects of these works to build new multi-agent based simulations dealing with this interesting and complex problem are discussed. The main four research lines in rumor dissemination found and discussed in this paper are: exploratory data analysis, rumor detection, epidemiological modeling, and multi-agent based social simulation. The survey shows that the reproducibility in the specialized literature has to be considerably improved. Finally, a free and open-source simulation tool implementing several of the models considered in this survey is presented.