Convolution Neural Networks (CNNs) are famous for their great performance in Computer Vision experiments achieving state-of-art results. However, recent works have shown that CNNs can perform well for Natural Language Processing. The whole idea consists of gathering the embeddings as an image. This paper presents the usage of the recently released Facebook fastText word embeddings as representation of word to perform the task of sentiment analysis. The interest in this work comes from the advent of social media and technological advances, which have been flooding the Internet with opinions. The results show that the proposed aproach outperforms the baseline models and it has similar performance to the state-of-art models.

Real-world engineering problems often require concurrent optimisation of several design objectives. These objectives are conflicting in most of the cases. Such an optimisation is generally called multi-objective or multi-criterion optimisation. Unlike single-objective optimisations, which target a single best solution, multi-objective optimisations aim at a set of equally good solutions. Classical multi-objective methods have serious drawbacks. The most limiting one is that one needs to apply the optimisation method, as many times as different solution are required. Evolutionary computations are now well-known. They are based on the Darwinian natural selection theory. They have been proven to be much more efficient than the classical methods as they can provide a whole set of good solutions after a single optimisation process. This paper introduces multi-criterion optimisation and states the classical multi-criterion optimisation problem. We review the most successful evolutionary algorithms for multi-objective optimisation. For each of the described methods, we sketch the underlying advantages vs. disadvantages. We give some statistics about applied work, and we survey the works about the most recent applications of the reviewed algorithms to solve real-world problems in science and technological fields.

Dynamic task allocation in a robotic swarm is a necessary process for proper management of the swarm. It allows the distribution of the identified tasks to be performed, among the swarm of robots, in such a way that a pre-defined proportion of execution of those tasks is achieved. In this context, there is no central unit to take care of the task allocation. So any algorithm proposal must be distributed, allowing every, and each robot in the swarm to identify the task it must perform. This paper proposes a distributed control algorithm to implement dynamic task allocation in a swarm robotics environment. The algorithm is inspired by the particle swarm optimization. In this context, each robot that integrates the swarm must run the algorithm periodically in order to control the underlying actions and decisions. The algorithm was implemented on ELISA III real swarm robots and extensively tested. The algorithm is effective and the corresponding performance is promising