Learning Extended Tree Augmented Naive Structures

This work proposes an extended version of the well-known tree-augmented naive Bayes (TAN) classifier where the structure learning step is performed without requiring features to be connected to the class. Based on a modification of Edmonds' algorithm, our structure learning procedure explores a superset of the structures that are considered by TAN, yet achieves global optimality of the learning score function in a very efficient way (quadratic in the number of features, the same complexity as learning TANs). We enhance our procedure with a new score function that only takes into account arcs that are relevant to predict the class, as well as an optimization over the equivalent sample size during learning. These ideas may be useful for structure learning of Bayesian networks in general. A range of experiments shows that we obtain models with better prediction accuracy than naive Bayes and TAN, and comparable to the accuracy of the state-of-the-art classifier averaged one-dependence estimator (AODE). We release our implementation of ETAN so that it can be easily installed and run within Weka.

Learning Bounded Tree-Width Bayesian Networks via Sampling

Learning Bayesian networks with bounded tree-width has attracted much attention recently, because low tree-width allows exact inference to be performed efficiently. Some existing methods [12, 14] tackle the problem by using k-trees to learn the optimal Bayesian network with tree-width up to k. In this paper, we propose a sampling method to efficiently find representative k-trees by introducing an Informative score function to characterize the quality of a k-tree. The proposed algorithm can efficiently learn a Bayesian network with tree-width at most k. Experiment results indicate that our approach is comparable with exact methods, but is much more computationally efficient.

Learning Bayesian Networks with Bounded Tree-Width via Guided Search

Bounding the tree-width of a Bayesian network can reduce the chance of overfitting, and allows exact inference to be performed efficiently. Several existing algorithms tackle the problem of learning bounded tree-width Bayesian networks by learning from k-trees as super-structures, but they do not scale to large domains and/or large tree-width. We propose a guided search algorithm to find k-trees with maximum Informative scores, which is a measure of quality for the k-tree in yielding good Bayesian networks. The algorithm achieves close to optimal performance compared to exact solutions in small domains, and can discover better networks than existing approximate methods can in large domains. It also provides an optimal elimination order of variables that guarantees small complexity for later runs of exact inference. Comparisons with well-known approaches in terms of learning and inference accuracy illustrate its capabilities.

Evaluation of the in vitro biological activities of extracts from carob tree and mediterranean oaks

Dissertação mest., Engenharia Biológica, Universidade do Algarve, 2009

Chemical and microbial changes during the natural fermentation of strawberry tree (Arbutus unedo L.) fruits

Aguardente de medronho is the name given in Portugal to a spirit made from the fermented fruit of Arbutus unedo (strawberry tree), a plant grown in the Mediterranean region. In order to gain a better understanding of the fermentation process, as it is performed in the farms, a natural fermentation with wild microbiota was carried out during 36 days, and some physicochemical and microbiological parameters were studied. The microbial parameters analyzed were total viable, lactic and acetic acids bacteria, and yeast counts. The physicochemical parameters monitored were sugars, minerals, ethanol, organic acids and pH. Yeasts were the main responsible for the fermentation of the fruits, as the lactic and acetic acids bacteria are absent. As the fermentation progressed, the sugars increased during the first 2 days and gradually decreased along the fermentation period. Maintaining the good quality of the product could contribute to the preservation and valorization of traditional resources that are of great importance to prevent their disappearance.

Studies on honey from the Algarve in view of its valorization

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

Entomofauna associated with the olive tree in southern Portugal

Since 2004 several studies have been carried out in order to identify the main insect species that usually inhabiting the olive ecosystem. The field trials have taken place in two olive groves, one situated in Olhão and the other one in Loulé, both in Algarve and also under Integrated Pest Management (IPM). The sampling techniques used differ according to their purpose (sticky traps, pheromone traps, pitfall traps and samples of aerial parts of the trees such as inflorescences, leaves, fruits and branches). Results showed that the main insect pests of olive tree in southern Portugal were the olive fruit fly Bactrocera oleae Gmelin (Diptera: Tephritidae) and the olive moth Prays oleae Bernard (Lepidoptera: Hyponeumetidae). Other insect pests were also found in our olive groves namely the olive psyllid Euphyllura olivina Costa (Homoptera: Psyllidae), the olive dark beetle Phloeotribus scarabaeoides Bernard (Coleoptera: Curculionidae), the mediterranean black scale Saissetia oleae (Olivier) (Homoptera: Coccidae) and the olive thrip Liothripes oleae Costa (Thysanoptera: Phlaeothripidae). Concerning the auxiliary insects that were found in our olives groves they belong to the following orders and families: Diptera (Syrphidae), Coleoptera (Carabidae, Coccinelidae and Staphylinidae), Hemiptera (Anthocoridae and Miridae), Neuroptera (Chrysopidae) and Hymenoptera (Braconidae, Encyrtidae, Eulophidae, Formicidae and Trichogrammatidae).

Yeast diversity in the Mediterranean strawberry tree (Arbutus unedo L.) fruits' fermentations

In the Mediterranean region the fruits of the strawberry tree (Arbutus unedo L.) may be fermented and distilled to produce a traditional beverage very much appreciated in Southern Europe. The aim of the present work was to study the diversity of the yeast population and the killer activity of the isolates identified as Saccharomyces cerevisiae, obtained during solid state industrial fermentations of the arbutus berries. The identification of the isolates was performed by the 5.8S rRNA-ITS region restriction analysis and by sequencing the D1/D2 region of the large subunit of the rRNA gene. At the start of the fermentations, various non-Saccharomyces species were detected including Aureobasidium pullulans, Dothichiza pithyophila, Dioszegia zsoltii, Hanseniaspora uvarum and yeasts belonging to the genera Metschnikowia, Cryptococcus and Rhodotorula. However, as the biological processes progressed the number of different species decreased with S. cerevisiae and Pichia membranaefaciens becoming dominant at advanced stages of the must fermentation that is characterized by high concentrations of ethanol. Forty three isolates identified as S. cerevisiae were tested for killer activity against two sensitive reference strains and Zygosaccharomyces bailii. Their killer sensitivity in relation to five killer referenced toxins (K2, K5, K8, K9 and K10) was also studied. Out of the isolates analyzed, 95.3% were sensitive and 4.7% were tolerant against the killer toxins tested. Only three isolates revealed killer activity against one sensitive strain and two of them against the spoiler yeast Z. bailii. The microbiota obtained revealed an interesting potential to be used as starter cultures to overcome unpredictable uncontrolled fermentations of the arbutus fruits as well as in other applications of biotechnological interest. (C) 2012 Elsevier Ltd. All rights reserved.

Enhanced bio-decolourisation of acid orange 7 by Shewanella oneidensis through co-metabolism in a microbial fuel cell

The decolourisation of acid orange 7 (AO7) (C.I.15510) through co-metabolism in a microbial fuel cell by Shewanella oneidensis strain 14063 was investigated with respect to the kinetics of decolourisation, extent of degradation and toxicity of biotransformation products. Rapid decolourisation of AO7 (>98% within 30 h) was achieved at all tested dye concentrations with concomitant power production. The aromatic amine degradation products were recalcitrant under tested conditions. The first-order kinetic constant of decolourisation (k) decreased from 0.709 ± 0.05 h−1 to 0.05 ± 0.01 h−1 (co-substrate – pyruvate) when the dye concentration was raised from 35 mg l−1 to 350 mg l−1. The use of unrefined co-substrates such as rapeseed cake, corn-steep liquor and molasses also indicated comparable or better AO7 decolourisation kinetic constant values. The fully decolourised solutions indicated increased toxicity as the initial AO7 concentration was increased. This work highlights the possibility of using microbial fuel cells to achieve high kinetic rates of AO7 decolourisation through co-metabolism with concomitant electricity production and could potentially be utilised as the initial step of a two stage anaerobic/aerobic process for azo dye biotreatment.

Implementação de uma marca de produção e distribuição de citrinos biológicos: Orange made in Algarve

Trabalho de projeto apresentado à Escola Superior de Comunicação Social como parte dos requisitos para obtenção de grau de mestre em Publicidade e Marketing.

LNT: a logical neighbor tree secure group communication scheme for wireless sensor networks

Secure group communication is a paradigm that primarily designates one-to-many communication security. The proposed works relevant to secure group communication have predominantly considered the whole network as being a single group managed by a central powerful node capable of supporting heavy communication, computation and storage cost. However, a typical Wireless Sensor Network (WSN) may contain several groups, and each one is maintained by a sensor node (the group controller) with constrained resources. Moreover, the previously proposed schemes require a multicast routing support to deliver the rekeying messages. Nevertheless, multicast routing can incur heavy storage and communication overheads in the case of a wireless sensor network. Due to these two major limitations, we have reckoned it necessary to propose a new secure group communication with a lightweight rekeying process. Our proposal overcomes the two limitations mentioned above, and can be applied to a homogeneous WSN with resource-constrained nodes with no need for a multicast routing support. Actually, the analysis and simulation results have clearly demonstrated that our scheme outperforms the previous well-known solutions.

Energy efficient scheduling for cluster-tree wireless sensor networks with time-bounded data flows: application to IEEE 802.15.4/ZigBee

Cluster scheduling and collision avoidance are crucial issues in large-scale cluster-tree Wireless Sensor Networks (WSNs). The paper presents a methodology that provides a Time Division Cluster Scheduling (TDCS) mechanism based on the cyclic extension of RCPS/TC (Resource Constrained Project Scheduling with Temporal Constraints) problem for a cluster-tree WSN, assuming bounded communication errors. The objective is to meet all end-to-end deadlines of a predefined set of time-bounded data flows while minimizing the energy consumption of the nodes by setting the TDCS period as long as possible. Sinceeach cluster is active only once during the period, the end-to-end delay of a given flow may span over several periods when there are the flows with opposite direction. The scheduling tool enables system designers to efficiently configure all required parameters of the IEEE 802.15.4/ZigBee beaconenabled cluster-tree WSNs in the network design time. The performance evaluation of thescheduling tool shows that the problems with dozens of nodes can be solved while using optimal solvers.

A time division beacon scheduling mechanism for IEEE 802.15.4/Zigbee cluster-tree wireless sensor networks

While the IEEE 802.15.4/Zigbee protocol stack is being considered as a promising technology for low-cost low-power Wireless Sensor Networks (WSNs), several issues in the standard specifications are still open. One of those ambiguous issues is how to build a synchronized multi-hop cluster-tree network, which is quite suitable for ensuring QoS support in WSNs. In fact, the current IEEE 802.15.4/Zigbee specifications restrict the synchronization in the beacon-enabled mode (by the generation of periodic beacon frames) to star-based networks, while it supports multi-hop networking using the peer-to-peer mesh topology, but with no synchronization. Even though both specifications mention the possible use of cluster-tree topologies, which combine multihop and synchronization features, the description on how to effectively construct such a network topology is missing. This paper tackles this problem, unveils the ambiguities regarding the use of the cluster-tree topology and proposes a synchronization mechanism based on Time Division Beacon Scheduling to construct cluster-tree WSNs. We also propose a methodology for an efficient duty cycle management in each router (cluster-head) of a cluster-tree WSN that ensures the fairest use of bandwidth resources. The feasibility of the proposal is clearly demonstrated through an experimental test bed based on our own implementation of the IEEE 802.15.4/Zigbee protocol.