André M. C. Campos

GNeurAge: An Evolutionary Agent-Based System for Classification Tasks

The use of intelligent agents in the structure of multiclassifier systems has been investigated in order to overcome some drawbacks of these systems and, as a consequence, to improve the performance of such systems. As a result of this, the NeurAge system was proposed. This system has presented good results in some centralized and distributed classification tasks. In this paper, an investigation of using evolutionary techniques in the functioning of the NeurAge (GNeurAge) is performed. In order to do this, we are going to use genetic algorithm in two different phases: in the choice of the initial classifier; and during the functioning of NeurAge (test phase).

How to address group dynamics in virtual worlds

The study of group dynamics highlights the activity in the group in terms of its performance and communication. The experience of facilitating virtual communities and teams (Eunice and Kimball in http://www.Tmn.com/~lisa/odn-teams.htm, 1997) suggests that groups go through the same stages either in face-to-face or in online mode. The paper brings together a theoretical framework based on the literature on virtual communities, Gestalt systems and online facilitation in order to address the issue of electronic togetherness, in particular from a group dynamics perspective. The empirical work on which the paper is based is an observation of a group of students in a training set playing a decision-making game. The model of Tuckman (Tuckman in Psychol Bull 63:384–399, 1965; Tuckman and Jensen in Group Organ Stud 2:419–427, 1977) is used as a framework within which to discuss the findings of the case. The paper finishes with concrete recommendations for facilitators of online communities and designers of the electronic spaces where these communities operate.

A Comparative Analysis of Data Distribution Methods in an Agent-Based Neural System for Classification Tasks

The NeurAge (Neural agents) system has been proposed as an alternative to transform the centralized decision making process of a multi-classifier system into a distributed, flexible and incremental one. This system has presented good results in some conventional (centralized) classification tasks. Nevertheless, in some classification tasks, relevant features might be distributed over a set of agent. These applications can be classified as distributed classification tasks. In this paper, a comparative investigation of the NeurAge system using some methods for data distribution will be performed. In addition, the performance of the NeurAge system will be compared with some existing multi-classifier systems.

Towards a Methodology for Developing Agent-Based Simulations: The MASim Methodology

This paper presents the general aspects of the MASim methodology, aimed for the development of agent-based simulations. MASim employs features common to the development of agent-based software as well as to the development of simulation models. It also borrows concepts used in mainstream of the software engineering process.

Using an Affective Attention Focus for Improving the Reasoning Process and Behavior of Intelligent Agents

Simulations based on cognitively rich agents can become a very intensive computing task, especially when the simulated world represents a complex system. Those simulations can however benefit from optimizations coming from the way in which agents react to changes in the simulated environment. This paper presents an approach for improving the efficiency of the decision-making process of autonomous agents in a simulation. The optimization is reached by dynamically adapting the agents perception to a bounded subset of all the agents surrounding elements, which contains only the most important elements for the agent at the current time. In other words, the agent is modeled as having a dynamic focus of attention.

Simulating working environments through the use of personality-based agents

This paper presents a multi-agent simulation system, named SimOrg (Simulation of human organization), in which the personality aspect is incorporated in the internal functioning of the agents. The general architecture of the agents has been based on the theory of human personality from Theodore Millon [13]. The presented simulation has been used to simulate working environments, focusing on the training process within an organization. The main aim of this paper is to analyze how personality can affect the performance of the agent when simulating working environments.

A flexible framework for representing personality in agents

Aiming to represent individuals in a very realistic way, several works have attempted to introduce personality characteristics into artificial agents, usually based on theories of human personality. However, different theories require different structures in the agent architecture, reflecting on the way they influence agent behavior. This paper presents a flexible framework to generalize the influence of personality traits on the behavior of the agents. The framework provides an easy way to test and compare different personality models. In order to validate the framework, a simulation of working environment was developed and three personality theories (Big Five, Jung and Millon) were tested.

From Individuals to Social and Vice-versa

The concept of cyclical influence between individuals and society is widely accepted, but hard to understand in all details. This paper proposes the use of three processes of social influence as a way to study the link between social and individual levels of abstraction. These processes are used to design an agent architecture which tries to provide explicit links to its social context. In order to detail the impact of the social influence, the architecture also includes personality and emotional aspects.

A personality-based model of agents for representing individuals in working organizations

This paper proposes an agent architecture which can be used to represent individuals within a working organization. The proposed architecture has been based on the theory of human personality and its working relationship from Theodore Milton. The main aim of this paper is to describe a suitable representation of individual behaviors which is able to be mapped to collective patterns of a human organization. The proposed architecture has been used in the SimOrg project, which aims to apply a multi-agent simulation in human organizations.

Introducing affective agents in recommendation systems based on relational data clustering

This paper proposes the use of a multi-agent system (MAS) with affective agents in a recommendation system based on relational data clustering. This MAS works as a mediator between the user and the data stored in the system. In the proposed system, after logging in, each user will have an affective agent, called Interface agent, for interaction purposes. This agent models the users data requests according to the users profile and its affective status, sending it to the Recommender agent, which recommends a set of map points to be visualized. The system analyzes the users feedback in order to verify whether the recommended information was satisfactory. This feedback is analyzed through the monitoring of the interaction interface.