Design of B-spline neural networks using a bacterial programming approach

The design phase of B-spline neural networks represents a very high computational task. For this purpose, heuristics have been developed, but have been shown to be dependent on the initial conditions employed. In this paper a new technique, Bacterial Programming, is proposed, whose principles are based on the replication of the microbial evolution phenomenon. The performance of this approach is illustrated and compared with existing alternatives.

Studies on honey from the Algarve in view of its valorization

Tese de dout., Química, Faculdade de Ciências e Tecnologia, Univ. do Algarve, 2012

Screening for biological activities with application in aquaculture in halophytes from the Algarve coastline

Infectious diseases often hamper the production of aquatic organisms in aquaculture systems, causing economical losses, environmental problems and consumer safety issues. The conventional way aquaculture producers had to control pathogens was by means of synthetic antibiotics and chemicals. This procedure had consequences in the emergence of more resilient pathogens, drug contamination of seafood products and local ecosystems. To avoid the repercussions of antibiotic use, vaccination has greatly replaced human drugs in western fish farms. However there is still massive unregulated antibiotic use in third world fish farms, so less expensive therapeutic alternatives for drugs are desperately needed. An alternative way to achieve disease control in aquaculture is by using natural bioactive organic compounds with antibiotic, antioxidant and/or immunostimulant properties. Such diverse biomolecules occur in bacteria, algae, fungi, higher plants and other organisms. Fatty acids, nucleotides, monosaccharides, polysaccharides, peptides, polyphenols and terpenoids, are examples of these substances. One promising source of bioactive compounds are salt tolerant plants. Halophytes have more molecular resources and defence mechanisms, when compared with other tracheophytes, to deal with the oxidative stresses of their habitat. Many halophytes have been used as a traditional food and medical supply, especially by African and Asian cultures. This scientific work evaluated the antibiotic, antioxidant, immunostimulant and metal chelating properties of Atriplex halimus L., Arthrocnemum macrostachyum Moric., Carpobrotus edulis L., Juncus acutus L. and Plantago coronopus L., from the Algarve coast. The antibiotic properties were tested against Listonella anguillarum, Photobacterium damselae piscicida and Vibrio fischeri. The immunostimulant properties were tested with cytochrome c and Griess assays on Sparus aurata head-kidney phagocytes. J. acutus ether extract inhibited the growth of P. damselae piscicida. A. macrostachyum, A. halimus, C. edulis, Juncus acutus and P. coronopus displayed antioxidant, copper chelating and iron chelating properties. These plants show potential as sources of bioactive compounds with application in aquaculture and in other fields.

Design of neuro-fuzzy models by evolutionary and gradient-based algorithms

All systems found in nature exhibit, with different degrees, a nonlinear behavior. To emulate this behavior, classical systems identification techniques use, typically, linear models, for mathematical simplicity. Models inspired by biological principles (artificial neural networks) and linguistically motivated (fuzzy systems), due to their universal approximation property, are becoming alternatives to classical mathematical models. In systems identification, the design of this type of models is an iterative process, requiring, among other steps, the need to identify the model structure, as well as the estimation of the model parameters. This thesis addresses the applicability of gradient-basis algorithms for the parameter estimation phase, and the use of evolutionary algorithms for model structure selection, for the design of neuro-fuzzy systems, i.e., models that offer the transparency property found in fuzzy systems, but use, for their design, algorithms introduced in the context of neural networks. A new methodology, based on the minimization of the integral of the error, and exploiting the parameter separability property typically found in neuro-fuzzy systems, is proposed for parameter estimation. A recent evolutionary technique (bacterial algorithms), based on the natural phenomenon of microbial evolution, is combined with genetic programming, and the resulting algorithm, bacterial programming, advocated for structure determination. Different versions of this evolutionary technique are combined with gradient-based algorithms, solving problems found in fuzzy and neuro-fuzzy design, namely incorporation of a-priori knowledge, gradient algorithms initialization and model complexity reduction.