Andrés Muñoz

Design and evaluation of an ambient assisted living system based on an argumentative multi-agent system

This paper focuses on ambient assisted living systems employed to monitor the ongoing situations of elderly people living independently. Such situations are represented here as contexts inferred by multiple software agents out of the data gathered from sensors within a home. Sensors can give an incomplete, sometimes ambiguous, picture of the world; hence, they often lead to inconsistent contexts and unreliability on the system as a whole. We report on a solution to this problem based on a multi-agent system where each agent is able to support its understanding of the context through arguments. These arguments can then be compared against each other to determine which agent provides the most reliable interpretation of the reality under observation.

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.

OntoSakai: On the optimization of a Learning Management System using semantics and user profiling

Abstract This paper proposes recommendation services and user profiling features in Learning Management Systems (LMS) by means of a semantic intelligent system combining context information and expert knowledge. LMS users’ context is represented through an ontology model called OntoSakai. It consists of four ontologies parceling different areas of the learning process: competences, users’ profiles, learning tools and semantic classification of the elements in an LMS. Thus, we provide a standardized common vocabulary about LMS elements and academic tasks developed within these platforms. This model also enables inference processes about the behavior of LMS users. Indeed, our system incorporates an extensible set of expert rules to offer recommendation and user profiling services. This combination of context information and expert knowledge could be easily integrated with other systems in the academic world in order to promote the interoperability between them. Specifically, in this paper we integrate our proposal into Sakai, a well-known LMS for university-level. As a result of this integration, OntoSakai is able to generate users’ profiles aimed at personalizing the use of LMS tools and to recommend resources to reach the optimum benefit in both lecturing and learning. As a proof of concept, a real case often detected in on-line students is shown as a running scenario where the services offered by OntoSakai could help them to improve their experiences and academic results.

ASBO: Argumentation System Based on Ontologies

Conflicts are unavoidable in open distributed systems. Belief or semantic conflicts are of special interest in multiagent systems, where agents need to communicate by exchanging knowledge. A common approach to deal with conflicts is the use of argumentation-based negotiation processes. There have been much work in the argumentation research arena. Amongst the outcomes of this research, some generic argumentation frameworks for handling inconsistences can be found, together with several persuasion dialogue systems. The goal of this paper is to contribute in advancing the state-of-art in argumentation by extending the basic mechanisms used in conventional argumentation frameworks. This contribution consists of a new and convenient style of attack to arguments and making explicit the argumentation process structure through an OWL-based ontology. Main benefits of this research are twofold. Firstly, the availability of a more realistic framework thanks to the definition of the new attack. Secondly, to enable automatic reasoning about the argumentation process itself. To illustrate this approach, we expose a persuasive argumentation scenario based on a real situation.Conflicts are unavoidable in open distributed systems. Belief or semantic conflicts are of special interest in multiagent systems, where agents need to communicate by exchanging knowledge. A common approach to deal with conflicts is the use of argumentation-based negotiation processes. There have been much work in the argumentation research arena. Amongst the outcomes of this research, some generic argumentation frameworks for handling inconsistences can be found, together with several persuasion dialogue systems. The goal of this paper is to contribute in advancing the state-of-art in argumentation by extending the basic mechanisms used in conventional argumentation frameworks. This contribution consists of a new and convenient style of attack to arguments and making explicit the argumentation process structure through an OWL-based ontology. Main benefits of this research are twofold. Firstly, the availability of a more realistic framework thanks to the definition of the new attack. Secondly, to enable automatic reasoning about the argumentation process itself. To illustrate this approach, we expose a persuasive argumentation scenario based on a real situation.

A context-aware solution for personalized en-route information through a P2P agent-based architecture

Communication technologies and the management of context aware information are two key researching lines in the intelligent transportation domain. A suited network infrastructure which allows Vehicle to Vehicle (V2V) and Vehicle to Infrastructure (V2I) communications, on one hand, and the provision of location-based information adapted to the driver, on the other hand, comprise two important components in vehicle telematics field. Following this idea, our work includes a communication infrastructure based on Peer to Peer (P2P) networks which deals with V2V and V2I requirements. Reasoning agents have been integrated in the designed network, so the vehicle and the infrastructures entities act as agents which adapt the information about the traffic area according to a driver profile. The paper describes also a prototype which implements both the vehicle edge, using an adapted vehicle, and the road edge, developing the network infrastructure and the knowledge management system. A performance study of the inference of contextual information has been added as well.

Intelligent Decision-Making for a Smart Home Environment with Multiple Occupants

Ambient Intelligence (AmI) systems rely on the use of environmental information to seamlessly assist users in their daily lives. As the development of AmI architectures is involving a larger amount of users and devices information, the potential occurrence of conflicts derived from the use of this information becomes higher. These conflicts are usually represented as inconsistencies between different pieces of information. As a result, the AmI systems need a mechanism to decide upon these conflicting situations in order to still give an appropriate response. In this chapter we propose the use of Argumentation as a methodology for automatically detecting and solving these conflicting situations. More specifically, we describe a smart TV service which combines different information about viewers’ presence and preferences and TV programs data to elaborate a list of recommending programs. Several arguments for and against recommending a same program are presented to the AmI system. Then, by means of argumentative techniques and several meta-decision criteria, the system is able to assist through mediation in the context of conflicting preferences or needs.

Heart health risk assessment system: a nonintrusive proposal using ontologies and expert rules.

According to the World Health Organization, the worlds leading cause of death is heart disease, with nearly two million deaths per year. Although some factors are not possible to change, there are some keys that help to prevent heart diseases. One of the most important keys is to keep an active daily life, with moderate exercise. However, deciding what a moderate exercise is or when a slightly abnormal heart rate value is a risk depends on the person and the activity. In this paper we propose a context-aware system that is able to determine the activity the person is performing in an unobtrusive way. Then, we have defined ontology to represent the available knowledge about the person (biometric data, fitness status, medical information, etc.) and her current activity (level of intensity, heart rate recommended for that activity, etc.). With such knowledge, a set of expert rules based on this ontology are involved in a reasoning process to infer levels of alerts or suggestions for the users when the intensity of the activity is detected as dangerous for her health. We show how this approach can be accomplished by using only everyday devices such as a smartphone and a smartwatch.

Developing an intelligent parking management application based on multi-agent systems and semantic web technologies

Multi-agent systems (MAS) are being adopted in multiple areas to deal with knowledge-based applications. On the other hand, Semantic Web technologies such as OWL and SWRL have shown to be useful in managing knowledge and reasoning about it. This paper proposes an architecture based on these technologies to develop an intelligent parking management application, where agents interact to reach a consensus about the assignment of a parking area to a vehicle. Moreover, this paper tackles the inherent problem related to the rise of conflicts in MAS by means of the integration of an argumentation system called ASBO (which is part of our previous work) into the proposed architecture.Multi-agent systems (MAS) are being adopted in multiple areas to deal with knowledge-based applications On the other hand, Semantic Web technologies such as OWL and SWRL have shown to be useful in managing knowledge and reasoning about it This paper proposes an architecture based on these technologies to develop an intelligent parking management application, where agents interact to reach a consensus about the assignment of a parking area to a vehicle Moreover, this paper tackles the inherent problem related to the rise of conflicts in MAS by means of the integration of an argumentation system called ASBO (which is part of our previous work) into the proposed architecture.

A survey on managing users' preferences in ambient intelligence

Understanding the importance of preference management in ambient intelligent environments is key to providing systems that are better prepared to meet users’ expectations. This survey provides an account of the various ways that preferences have been handled in Artificial Intelligence. Our analysis indicates that most of those techniques lack the ability to handle ambiguity and the evolution of preferences over time. Further exploration shows that argumentation can provide a feasible solution to complement existing work. We illustrate our claim by using an intelligent environment case study.