All the eighteen Frankia strains isolated from ten different actinorhizal host plants showed uptake hydrogenase activity. The activity of this enzyme is further increased by addition of nickel. Nickel also enhanced the degree of hydrogenase transfer into the membranes of Frankia, indicating the role of this metal in the processing of this enzyme. The uptake hydrogenase of Frankia is most probably a Ni-Fe hydrogenase. Genome characterization revealed the presence of two hydrogenase genes (syntons) in Frankia, which are distinctively separated in all the three available Frankia genomes. Both hydrogenase syntons are also commonly found in other Frankia strains. The structural, regulatory and accessory genes of both hydrogenase synton #1 and #2 are arranged closely together, but in a clearly contrasting organization. Hydrogenase synton #1 and #2 of Frankia are phylogenetically divergent and that hydrogenase synton #1 is probably ancestral among the actinobacteria. Hydrogenase synton #1 (or synton #2) of Frankia sp. CcI3 and F. alni ACN14a are similar in gene arrangement, content and orientation, while the syntons are both reduced and rearranged in Frankia sp. EANpec. The hydrogenases of Frankia sp. CcI3 and F. alni ACN14a are phylogenetically grouped together but never with the Frankia sp. EAN1pec, which is more closely related to the non-Frankia bacteria than Frankia itself. The tree topology is indicative of a probable gene transfer to or from Frankia that occurred before the emergence of Frankia. All of the available evidence points to hydrogenase gene duplication having occurred long before development of the three Frankia lineages. The uptake hydrogenase synton #1 of Frankia is more expressed under free-living conditions whereas hydrogenases synton #2 is mainly involved in symbiotic interactions. The uptake hydrogenase of Frankia can also be manipulated to play a larger role in increasing the efficiency of nitrogen fixation in the root nodules of the host plants, there by minimizing the need for environmentally unfriendly and costly fertilizers. The hydrogen-evolving hydrogenase activity was recorded in only four Frankia strains: F. alni UGL011101, UGL140102, Frankia sp. CcI3 and R43. After addition of 15mM Nicl2, activity was also detected in F. alni UGL011103, Frankia sp. UGL020602, UGL020603 and 013105. Nickel also increased the activity of hydrogen-evolving hydrogenases in Frankia, indicating that Frankia may have different types of hydrogen-evolving hydrogenases, or that the hydrogen-evolving hydrogenases may at least be regulated differently in different Frankia strains. The fact that Frankia can produce hydrogen is reported only recently. The knowledge of the molecular biology of Frankia hydrogenase is, therefore, of a paramount importance to optimize the system in favor of hydrogen production. Frankia is an attractive candidate in search for an organism efficient in biological hydrogen production since it can produce a considerable amount of hydrogen.

Ecological modelling problems have characteristics both featured in other modelling fields and specific ones, hence, methods developed and tested in other research areas may not be suitable for modelling ecological problems or may perform poorly when used on ecological data. This thesis identifies issues associated with the techniques typically used for solving ecological problems and develops new generic methods for decision support, especially suitable for ecological data modelling, which are characterised by: (1) adaptive learning, (2) knowledge discovery and (3) accurate prediction. These new methods have been successfully applied to challenging real world ecological problems. Despite the fact that the number of possible applications of computational intelligence methods in ecology is vast, this thesis primarily concentrates on two problems: (1) species establishment prediction and (2) environmental monitoring. Our review of recent papers suggests that multi-layer perceptron networks trained using the backpropagation algorithm are most widely used of all artificial neural networks for forecasting pest insect invasions. While the multi-layer perceptron networks are appropriate for modelling complex nonlinear relationships, they have rather limited exploratory capabilities and are difficult to adapt to dynamically changing data. In this thesis an approach that addresses these limitations is proposed. We found that environmental monitoring applications could benefit from having an intelligent taste recognition system possibly embedded in an autonomous robot. Hence, this thesis reviews the current knowledge on taste recognition and proposes a biologically inspired artificial model of taste recognition based on biologically plausible spiking neurons. The model is dynamic and is capable of learning new tastants as they become available. Furthermore, the model builds a knowledge base that can be extracted during or after the learning process in form of IF-THEN fuzzy rules. It also comprises a layer that simulates the influence of taste receptor cells on the activity of their adjacent cells. These features increase the biological relevance of the model compared to other current taste recognition models. The proposed model was implemented in software on a single personal computer and in hardware on an Altera FPGA chip. Both implementations were applied to two real-world taste datasets.In addition, for the first time the applicability of transductive reasoning for forecasting the establishment potential of pest insects into new locations was investigated. For this purpose four types of predictive models, built using inductive and transductive reasoning, were used for predicting the distributions of three pest insects. The models were evaluated in terms of their predictive accuracy and their ability to discover patterns in the modelling data. The results obtained indicate that evolving connectionist systems can be successfully used for building predictive distribution models and environmental monitoring systems. The features available in the proposed dynamic systems, such as on-line learning and knowledge discovery, are needed to improve our knowledge of the species distributions. This work laid down the foundation for a number of interesting future projects in the field of ecological modelling, robotics, pervasive computing and pattern recognition that can be undertaken separately or in sequence.

This thesis proposes a novel feature selection and classification method employing evolving spiking neural networks (eSNN) and evolutionary algorithms (EA). The method is named the Quantum-inspired Spiking Neural Network (QiSNN) framework. QiSNN represents an integrated wrapper approach. An evolutionary process evolves appropriate feature subsets for a given classification task and simultaneously optimises the neural and learning-related parameters of the network. Unlike other methods, the connection weights of this network are determined by a fast one-pass learning algorithm which dramatically reduces the training time. In its core, QiSNN employs the Thorpe neural model that allows the efficient simulation of even large networks. In QiSNN, the presence or absence of features is represented by a string of concatenated bits, while the parameters of the neural network are continuous. For the exploration of these two entirely different search spaces, a novel Estimation of Distribution Algorithm (EDA) is developed. The method maintains a population of probabilistic models specialised for the optimisation of either binary, continuous or heterogeneous search spaces while utilising a small and intuitive set of parameters. The EDA extends the Quantum-inspired Evolutionary Algorithm (QEA) proposed by Han and Kim (2002) and was named the Heterogeneous Hierarchical Model EDA (hHM-EDA). The algorithm is compared to numerous contemporary optimisation methods and studied in terms of convergence speed, solution quality and robustness in noisy search spaces. The thesis investigates the functioning and the characteristics of QiSNN using both synthetic feature selection benchmarks and a real-world case study on ecological modelling. By evolving suitable feature subsets, QiSNN significantly enhances the classification accuracy of eSNN. Compared to numerous other feature selection techniques, like the wrapper-based Multilayer Perceptron (MLP) and the Naive Bayesian Classifier (NBC), QiSNN demonstrates a competitive classification and feature selection performance while requiring comparatively low computational costs.

This thesis proposes a novel feature selection and classification method employing evolving spiking neural networks (eSNN) and evolutionary algorithms (EA). The method is named the Quantum-inspired Spiking Neural Network (QiSNN) framework. QiSNN represents an integrated wrapper approach. An evolutionary process evolves appropriate feature subsets for a given classification task and simultaneously optimises the neural and learning-related parameters of the network. Unlike other methods, the connection weights of this network are determined by a fast one-pass learning algorithm which dramatically reduces the training time. In its core, QiSNN employs the Thorpe neural model that allows the efficient simulation of even large networks. In QiSNN, the presence or absence of features is represented by a string of concatenated bits, while the parameters of the neural network are continuous. For the exploration of these two entirely different search spaces, a novel Estimation of Distribution Algorithm (EDA) is developed. The method maintains a population of probabilistic models specialised for the optimisation of either binary, continuous or heterogeneous search spaces while utilising a small and intuitive set of parameters. The EDA extends the Quantum-inspired Evolutionary Algorithm (QEA) proposed by Han and Kim (2002) and was named the Heterogeneous Hierarchical Model EDA (hHM-EDA). The algorithm is compared to numerous contemporary optimisation methods and studied in terms of convergence speed, solution quality and robustness in noisy search spaces. The thesis investigates the functioning and the characteristics of QiSNN using both synthetic feature selection benchmarks and a real-world case study on ecological modelling. By evolving suitable feature subsets, QiSNN significantly enhances the classification accuracy of eSNN. Compared to numerous other feature selection techniques, like the wrapper-based Multilayer Perceptron (MLP) and the Naive Bayesian Classifier (NBC), QiSNN demonstrates a competitive classification and feature selection performance while requiring comparatively low computational costs.

In marine environments, eukaryotic marine microalgae coexist with the viruses that infect them. Marine microalgae are the main primary producers in the oceans and are at the base of the marine food web. Viruses play important roles in top-down control of algae populations, cycling of organic matter, and as evolutionary drivers of their hosts. Algae must adapt in response to the strong selection pressure that viruses impose for resistance to infection. In addition to biotic selection pressures such as viral infections, algae must also adapt to their abiotic environment. Global climate change is affecting temperature, salinity, pH, light and nutrient concentrations in the oceans, particularly in surface waters, where microalgae live. Currently, little is known about how consistent the effects of viruses on their hosts are, whether the cost of host resistance varies across environments, and whether there is a trade-off between maintaining resistance to viruses and adapting to other environmental changes. The marine picoeukaryote Ostreococcus tauri is abundant in Mediterranean lagoons, where it experiences large fluctuations in environmental conditions and co-occurs with lytic viruses (Ostreococcus tauri viruses – OtVs). Viral infection causes lysis of susceptible (S) cells, however a small proportion of cells are resistant (R) and avoid lysis. Some resistant O. tauri populations can coexist with infectious viruses, and it has been proposed that these viruses are produced by a minority of susceptible cells within a mainly resistant population. These populations are referred to as resistant producers (RP). Virus production in RP populations is unstable and eventually they shift to R populations. I used O. tauri and one of its viruses, OtV5, as a model system to investigate whether cells that are susceptible or resistant to virus infection adapt to environmental change differently and whether there is a cost of being resistant. For the first time, I evolved susceptible and resistant hosts of a marine alga separately under a range of environments and directly compared their plastic and evolved responses. I showed that resistant populations of O. tauri maintained their resistance for more than 200 generations in the absence of viruses across all environments, indicating that the resistance mechanism is difficult to reverse. Furthermore, I did not detect a cost of being resistant, as measured by population growth rate and competitive ability. Virus production in RP populations stopped in all environments and all populations became R. In addition, I found that virus production in RP O. tauri populations can fluctuate before completely ceasing, and that phosphate affected the length of time it took for virus production to stop. These results, combined with mathematical modelling of O. tauri infection dynamics, provide support for the prediction that RP populations consist of a mixed population of susceptible and resistant cells. By examining multiple environments and resistance types, we can better understand first, how microalgae populations adapt to environmental change and second, the ecological and evolutionary consequences of maintaining resistance to viruses in common marine picoeukaryotes.

I denna studie analyserar vi numeriska metoder som beskriver koncentrationen av ytaktiva ämnen (surfaktanter). Dessa surfaktanter befinner sig i olösliga vätskor som separeras av ett tidsberoende gränssnitt som påverkas av ett givet hastighetsfält. Surfaktanter har stor inverkan på vätskesystem på grund av deras förmåga att sänka ytspänningen mellan två vätskor, exempelvis kan dessa användas för att göra olja mer lösbart i vatten. En vanlig strategi vid implementering av finita elementmetoder (FEM) för att lösa liknande problem är att låta beräkningsnätet anpassas efter den tidsberoende domänen, vilket kräver omdiskretisering i varje tidssteg. Således har så kallade oanpassade metoder, som inte kräver att beräkningsnätet anpassas efter domänen, blivit ett användbart alternativ till standard FEM för komplicerade tidsberoende problem. Oanpassade metoder använder ett kontant bakgrundsnät som täcker beräkningsdomänen i varje tidssteg. I denna studie tillämpar vi skurna finita elementmetoder (CutFEM) på två olika matematiska modeller av surfaktanterna. I den första modellen betraktas endast surfaktantkoncentrationen i bulkgeometrin, där koncentrationen ges av en konvektion-diffusionsekvation. I den andra modellen löses istället två ekvationer som är kopplade till varandra (en i bulkdomänen och en på ytan) genom en icke-linjär kopplingsmodell.

No description available.

No description available.

Yes / Data streams have arisen as a relevant topic during the last few years as an efficient method for extracting knowledge from big data. In the robust layered ensemble model (RLEM) proposed in this paper for short-term traffic flow forecasting, incoming traffic flow data of all connected road links are organized in chunks corresponding to an optimal time lag. The RLEM model is composed of two layers. In the first layer, we cluster the chunks by using the Graded Possibilistic c-Means method. The second layer is made up by an ensemble of forecasters, each of them trained for short-term traffic flow forecasting on the chunks belonging to a specific cluster. In the operational phase, as a new chunk of traffic flow data presented as input to the RLEM, its memberships to all clusters are evaluated, and if it is not recognized as an outlier, the outputs of all forecasters are combined in an ensemble, obtaining in this a way a forecasting of traffic flow for a short-term time horizon. The proposed RLEM model is evaluated on a synthetic data set, on a traffic flow data simulator and on two real-world traffic flow data sets. The model gives an accurate forecasting of the traffic flow rates with outlier detection and shows a good adaptation to non-stationary traffic regimes. Given its characteristics of outlier detection, accuracy, and robustness, RLEM can be fruitfully integrated in traffic flow management systems.

Titel: Employer branding i matchningsprocesser - En kvalitativ studie om vilken påverkan employer branding har på matchningsprocessen mellan arbetssökande och arbetsgivare Syfte: Studien syftar till att inventera och beskriva vilken påverkan employer branding har på matchningsprocessen mellan arbetssökande och arbetsgivare vid rekrytering, med utgångspunkt i vad arbetssökande värderar hos arbetsgivare samt vad arbetsgivare värderar hos arbetssökande. Metod: Studien är baserad på en deduktiv forskningsansats med en kvalitativ forskningsstrategi. Vidare tillämpar studien en tvärsnittsdesign med semistrukturerade intervjuer som datainsamlingsmetod. Studien tar avstamp i en etablerad teoretisk grund, vilket har resulterat i identifiering av fyra huvudområden: employer branding, rekrytering, matchningsprocess och arbetssökandes förväntningar på arbetsgivare. De fyra huvudområdena har sedan legat till grund vid utformningen av intervjuguiden. Slutsats: Studiens resultat visar att employer branding har en betydande roll för att uppnå en precis matchning mellan arbetsgivare och arbetssökande. Genom att tydligt kommunicera organisationens identitet kan arbetsgivare attrahera och behålla högkvalificerade talanger. Studien har även resulterat i att en kontinuerlig anpassning av organisationens strategier kring employer branding är nödvändig för att möta förändrade förväntningar på arbetsmarknaden. Ett starkt employer brand differentierar organisationer och bidrar till att säkerställa en optimal matchning mellan arbetsgivare och arbetssökande. / Title: Employer Branding in Matchmaking Processes - A qualitative study on the impact of employer branding on the matching process between job applicants and employers Purpose: This study endeavors to inventory and describe the influence of employer branding on the matchmaking process between job applicants and employers during recruitment, with a focus on the attributes valued by job applicants and employers alike. Method: Employing a deductive research approach and a qualitative research strategy, this study employs a cross-sectional design utilizing semi-structured interviews as the primary data collection method. Rooted in an established theoretical framework, the study identifies four principal domains: employer branding, recruitment, matchmaking processes, and job applicants' expectations of employers. These domains underpin the construction of the interview guide. Conclusion: The findings underscore the pivotal role of employer branding in facilitating precise alignment between employers and job applicants. By articulating the organization's identity, employers can effectively attract and retain highly skilled talents. Additionally, the study underscores the imperative of continually adapting organizational employer branding strategies to address evolving expectations in the labor market. A robust employer brand serves to differentiate organizations and fosters optimal alignment between employers and job applicants.