Integration of algorithms investment theory and artificial intelligence allows one to create a support system for investors in exchange markets based on the ensemble long-short-term-memory (LSTM) recurrent neural networks (RNN). The proposed contains five stages: preparation historical data, prediction by an LSTM RNNs, assessment distributions, portfolio formation verification. process outputs multi-modal distribution, which provides useful information investors. research compares four...

Considering the great number of agent-oriented methodologies that can be found in literature, and fact each one defines its own concepts system structure, main challenges software engineering (AOSE) research is how to make these interoperable. By defining used a specific domain non-ambiguous way, meta-modelling may represent step towards such interoperability. Consequently objective AOSE TFG (Technical Forum Group) establish strategy for identifying common meta-model could widely adopted by...

The agent oriented approach is doing great steps towards its (not yet reached) maturity; from a software engineering point of view, it today positively used for the analysis and design complex systems. In this paper, which related to activity AgentLink AOSE TFG (Agent Oriented Software Engineering Technical Forum Group), we provide perspective current research trends in area with specific attention results coming European groups. We start discussion what are agents, specially engineer....

AbstractAgent-based modelling and simulation (ABMS) had an increasing attention during the last decade. However, weak validation verification of agent-based models makes ABMS hard to trust. There is no comprehensive tool set for models, which demonstrates that inaccuracies exist and/or reveals existing errors in model. Moreover, on practical side, many frameworks are use. In this sense, we designed developed a generic testing framework conduct models. This paper presents our detail its...

Understanding functional synaptic connectivity of human central nervous system is one the holy grails neuroscience. Due to complexity system, it common reduce problem smaller networks such as motor unit pathways. In this sense, we designed and developed a simulation model that learns acting in same way single units by using findings on subjects. The based self-organizing agents whose nominal cooperative behaviors are current knowledge biological neural networks. results show generates...

In order to achieve an optimal learning in a serious game, continuously adapted game experience is needed. This paper presents multi-agent system which dynamically control the orientation of according players' actions. Irrespective application domain, scenario abstracted and elements composing it are identified. These then encapsulated into cooperative agents interact collectively adapt they related to. The feasibility this approach shown by applying real scenario.

Most frail older persons are living at home, and we face difficulties in achieving seamless monitoring to detect adverse health changes. Even more important, this lack of follow-up could have a negative impact on the choices made by individuals their care partners. People give up homes for reassuring environment medicalized facility. We developed low-cost unobtrusive sensor-based solution trigger automatic alerts case an acute event or subtle changes over time. It facilitate adults' own...

Control of game parameters to reach domain-related objectives cannot be easily handled with classic control theory approaches. Given the dynamics and complexity modern engines, diversity human players their constantly changing nature, this paper advocates for means tune in real time, no use or users models. The proposed approach, based on a multi-agent system, is used two dynamic systems before analyzing results.

ADELFE methodology is dedicated to applications characterized by openness and the need of system adaptation an environment. These are designed using AMAS theory [2]. Self-organization founded on capacity agent possesses be locally “cooperative”, i.e. it able recognize cooperation failures called “Non Cooperative Situations” (NCS, which could related exceptions in classical programs) treat them. It based RUP process follows its main phases: preliminary requirements, final analysis design,...

In a context of rapidly growing population elderly people, this paper introduces novel method for behavioural anomaly detection relying on self-organized learning process. This first models the Circadian Activity Rhythm set sensors and compares it to nominal profile determine variations in patients' activities. The anomalies are detected by multi-agent system as linear relation those variations, weighted influence parameters. problem adaptation particular patient then becomes adequate Those...