News

The paper An Agent-based Simulation Model for Emergency Egress, by Álvaro Carrera, Eduardo Merino, Pablo Aznar, Guillermo Fernández y Carlos A. Iglesias, has been presented at the 15th International Conference on Distributed Computing and Artificial Intelligence (DCAI 2018) in a Special Session on Social Modelling of Ambient Intelligence in Large Facilities (SMAILF 2018).

Abstract. Unfortunately, news regarding tragedies involving crowd evacuations are becoming more and more common. Understanding disasters and crowd emergency evacuation behaviour is essential to define effective evacuation protocols. This paper proposes an agent-based model of egress behaviour consisting of three complementary models: (i) model of people moving in a building in normal circumstances, (ii) policies of egress evacuation, and (iii) social models for integrating models (e.g. affiliation) that explain the social behaviour and help in mass evacuations. The proposed egress model has been evaluated in a university building and the results show how these models can help to better understand egress behaviour and apply this knowledge for improving the design and execution evacuation plans.

The DCAI 2018 conference was held June 20-22, at Toledo, Spain.

Cuándo:  martes 19 de junio a las 16:00

Dónde: sala B-225 de la ETSIT (UPM).

Ponente: Jürgen Dunkel

Título: Learning Complex Event Processing Rules with Genetic Programming

Resumen: Complex Event Processing (CEP) is an established software technology to extract relevant information from massive data streams. Currently, domain experts have to determine manually CEP rules that define a situation of interest. However, often CEP rules cannot be formulated by experts, because the relevant interdependencies and relations between the data are not explicitly known, but inherently hidden in the data streams. To cope with this problem, we present a new learning approach for CEP rules, which is based on Genetic Programming. We discuss in detail the different building blocks of Genetic Programming and how to adjust them to CEP rule learning. The extensive experiments, with synthetic data as well a real world data, show, that Genetic Programming has a high potential for learning CEP rules. In most of our experiments we could derive CEP rules with a nearly perfect recall and precision. 

During 13th and 14th of June, Trivalent consortium is holding the 1st year meeting in Rome. The main goals of this meeting  are: i)  sharing the progress of every task and work package; ii)  look for collaborations across work packages and iii) plan and schedule next steps

The main results of GSI are published here: http://www.gsi.dit.upm.es/index.php/es/trivalent

Este viernes 15 de junio se leen varios trabajos fin de titulación en la B-225, estáis todos invitados. Si estáis interesados, podéis encontrar las memorias y vídeos en la web.

The paper Towards an Autonomic Bayesian Fault Diagnosis Service for SDN Environments based on a Big Data Infrastructure, by Fernando Benayas, Álvaro Carrera and Carlos A. Iglesias, has been presented at the The Fifth IEEE International Conference on Software Defined Systems (SDS-2018).

The SDS 2018 aims to investigate the opportunities and in all aspects of Software Defined Systems (SDS). In addition, it seeks for novel contributions that help mitigating SDS challenges. That is, the objective of SDS 2018 is to provide a forum for scientists, engineers, and researchers to discuss and exchange new ideas, novel results and experience on all aspects of Software Defined Systems.

Abstract. Software Defined Networks (SDN) are gaining momentum as a solution for current and future networking issues. Its programmability and centralised control enables a more dynamic management of the network. But this feature introduces the cost of a potential increase in failures, since every modification introduced on the control plane is a new possibility for failures to appear and cause a decrement of the quality for the offered service. Following a classical approach, this kind of problems could be solved increasing the number of high skilled human operators, which would dramatically increase network operation cost. Our approach is to apply Machine Learning and Data Analysis for monitoring and diagnosis SDN networks with the goal of automating these tasks. In this paper, we present an architecture for a self-diagnosis service which is deployed on top of a SDN management platform. In addition, a prototype of the proposed service with different diagnosis models for SDN networks has been developed. The evaluation shows encouraging results which will be explored in future works.

The SDS-2018 conference was held April 23-26, at Barcelona, Spain.