We consider the problem of extracting a specific feature from a noisy signal generated by a multi-channels Remote Field Eddy Current Sensor. The sensor is installed on a mobile robot whose mission is the detection of anomalous regions in metal pipelines. Given the presence of noise that characterizes the data series, anomaly signals could be masked by noise and therefore difficult to identify in some instances. In order to enhance signal peaks that potentially identify anomalies we consider an entropy filter built on a-posteriori probability density functions associated with data series. Thresholds based on the Neyman-Pearson criterion for hypothesis testing are derived. The algorithmic tool is applied to the analysis of data from a portion of pipeline with a set of anomalies introduced at predetermined locations. Critical areas identifying anomalies capture the set of damaged locations, demonstrating the effectiveness of the filter in detection with Remote Field Eddy Current Sensor.

Empirical Bayes methods have been around for a long time and have a wide range of applications. These methods provide a way in which historical data can be aggregated to provide estimates of the posterior mean. This thesis revisits some of the empirical Bayesian methods and develops new applications. We first look at a linear empirical Bayes estimator and apply it on ranking and symbolic data. Next, we consider Tweedie’s formula and show how it can be applied to analyze a microarray dataset. The application of the formula is simplified with the Pearson system of distributions. Saddlepoint approximations enable us to generalize several results in this direction. The results show that the proposed methods perform well in applications to real data sets.

With the shift toward online shopping, it has become necessary to customize customers' needs and give them more choices. Before making a purchase, buyers research the products' features. The recommender systems facilitate the search task for customers by narrowing down the search space within specific products that align with the customer's needs. A recommender system uses clustering to filter information, calculating the similarity between members of a cluster to determine the factors that will lead to more accurate predictions. We propose a new method for predicting scores in machine learning recommender systems using the Gaussian mixture model clustering and the Pearson correlation coefficient. The proposed method is applied to MovieLens data. The results are then compared to three commonly used methods: Pearson correlation coefficients, K-means, and fuzzy C-means algorithms. As a result of increasing the number of neighbors, our method shows a lower error than others. Additionally, the results depict that accuracy will increase as the number of users increases. Our model, for instance, is 5% more accurate than existing methods when the neighbor size is 30. Gaussian mixture clustering chooses similar users and takes into account the scores distance when choosing nearby users that are similar to the active user.

No description available.

Recently, research has begun to pay more attention to the cognitive functions associated with auditory perception. In this study, two tests are performed to investigate the correlation between the ability to recognise speech-in-noise and the performance of sensory memory, as well as to investigate whether the performance would improve during the sensory memory test. For measuring speech-in-noise, the Hagerman test was used. A random noise test to detect deviant noises was used to measure sensory memory. In total 16 participantstook part in the study (mean age=24.8125, SD=3.14), half of the group began with the Hagerman test, and the other half with the random noise test. Two different statistical analyses were performed on the data. For examining the correlation between the performance on the Hagerman test and the random noise test, a Pearson correlation was used. The results were as follows: p = 0.4962, r = -0.1835734. Observing the results, the tests did indicate a slight negative correlation regarding the r-value, but not a significant correlation. Thus, the analysis did not derive any significant results. The second analysis was a dependent t-test to examine whether there was an improvement in performance during the random noise test, as it was divided into four separate blocks. The analysis showed the following results: t = 1.0266, df = 28.943, p = 0.3131. These results were not significant, though observing the block graph might indicate a tendency for improvement. For further studies, the random noise test should perhaps be modified into an easier version. This is based upon the data, as many of the participants merely got a score above, or even below, chance. Further studiesshould also use a higher number of participants as well to increase the chance of receiving significant results from the tests.

We define and study multivariate and matrix variate skew Pearson type VII and skew t-distributions. We derive the marginal and conditional distributions, the linear transformation, and the stochastic representations of the multivariate and matrix variate skew Pearson type VII distributions and skew t-distributions. Also, we study the limiting distributions.

No description available.

No description available.

In the fall of 2006, soils from three small valleys in the Laxemar/Oskarshamn area were sampled. A total of eight composite samples were characterized for a number of soil parameters that are important for geochemical sorption and were later also used in batch sorption experiments. Solid/liquid partition coefficients (Kd values) were then determined for seven radionuclides in each of the eight samples. To contribute to the interpretation of the sorption results together with the soil characterizations, this study aims to describe the sorption behavior of the radionuclides caesium, nickel and uranium and also discern which parameters that could provide a basis for estimating the strength of sorption of radionuclides in general. The methodology included quantitative methodologies such as compilation of chemical equilibrium diagrams by the software Hydra/Medusa and correlation analyses using the statistical software SPSS statistics. Based on the speciation diagrams of each radionuclide and identified important linear and non-linear relationships of the Kd values with a number of soil parameters, the following soil- and soil solution properties were found to have controlled the sorption of Cs, Ni and U, respectively, in the Laxemar soils. Cs: the specific surface area of the soil coupled to the clay content. Ni: the cation exchange capacity, alkaline solution pH, soil organic matter and dissolved organic matter. U: the cation exchange capacity, soil organic matter, dissolved organic matter, dissolved carbonate and alkaline solution pH. The soil that showed the strongest sorption varied between the nuclides, which can be related to the individual sorption behavior of caesium, nickel and uranium, as well as the different physicochemical properties of the soils. The parameters that should be prioritized in characterizations of soil samples are identified to be: solution pH, the cation exchange capacity, the specific surface area of the soil, soil organic matter and soil texture (clay content). / För att kunna fatta beslut relaterade till hypotetisk framtida kontaminering från slutförvar av radioaktivt avfall är det direkt avgörande att förstå mobiliteten av radioaktiva element i miljön. Sorption är en av de viktigaste kemiska mekanismerna som kan minska spridningen av radionuklider i vatten/jord/bergssystem, där nukliderna fördelar sig mellan vätskefasen och ytor på fasta partiklar i dessa system. Fördelningskoefficienter (Kd värden) används generellt som ett kvantitativt mått på sorptionen, där ett högt Kd värde innebär att en större andel av ämnet i fråga är bundet till den fasta fasen. Under hösten 2006 togs jordprover från tre dalgångar i Laxemar/Oskarshamn. Totalt åtta jordprover karakteriserades för ett antal jordparametrar som är viktiga för geokemisk sorption och användes senare i batchförsök tillsammans med ett naturligt grundvatten. Fördelningskoefficienter (Kd värden) bestämdes för sju radionuklider (Cs, Eu, I, Ni, Np, Sr and U) för vart och ett av de åtta jordproverna. För att bidra till tolkningen av sorptionsresultaten tillsammans med jordprovernas egenskaper syftar denna studie till att beskriva sorptionsbeteendet hos radionukliderna caesium, nickel och uran samt urskilja vilka parametrar som kan fungera som grund för att uppskatta sorptionsstyrkan av radionuklider i allmänhet. För att uppnå detta syfte så har studien följande mål. Identifiera de jord- och marklösningsegenskaper som kontrollerar sorptionen av Cs, Ni respektive U i de åtta Laxemar proverna. Bestämma vilket Laxemar-jordprov som starkast sorberar de tre radionukliderna. Identifiera de jordparametrar som bör prioriteras vid jordkarakteriseringar, baserat på deras sorptionsinflytande, för att kunna uppskatta Kd värden endast med begränsad information om ett jordsystem. Metoden innefattade kvantitativa metoder såsom sammanställning av kemiska jämviktsdiagram med programvaran Hydra/Medusa och korrelationsanalyser med hjälp av statistikprogramvaran SPSS statistics. De kemiska jämviktsdiagrammen bidrog till att beskriva specieringen av respektive nuklid som en funktion av pH och korrelationsanalyserna bidrog till att identifiera linjära samband mellan par av variabler, tex mellan Kd och jordparametrar. Baserat på specieringsdiagrammen för varje radionuklid och identifierade viktiga linjära och icke-linjära förhållanden mellan Kd-värdena och ett antal jordparametrar har följande egenskaper hos jordarna och marklösningen visat sig huvudsakligen kontrollera sorptionen av Cs, Ni respektive U i de åtta Laxemar jordarna: För caesium gäller jordens specifika ytarea kopplad till lerinnehållet, medan för nickel är det katjonbytarkapaciteten, organiskt material, alkaliska pH-värden samt löst organiskt material. Sorptionen av uran befanns kontrolleras av katjonbytarkapaciteten, organiskt material, löst organiskt material, alkaliska pH-värden samt lösta karbonater. Den jord som visade starkast sorption varierar mellan de tre nukliderna, vilket kan relateras till nuklidernas individuella sorptionsbeteende i jord samt jordarnas olika fysikaliska och kemiska egenskaper. Parametrarna som bör prioriteras vid karaktärisering av jordprov identifierades vara: pH, katjonbytarkapaciteten, jordens specifika ytarea, mängden organiskt material samt jordtexturen (lerinnehåll).

The purpose with this study is to show how the subjective uncertainty analysis can be applied to the cost estimates and examine how the attitudes towards risk budgeting of construction projects works. Biogas Mellannorrland is based on a case study and focuses on four location options for the biogas plant. For each option, a cost estimate in form of a fixed cost and variable cost elements, where the latter are uncertain. Pearson-Tukey method have been applied as a method to estimating subjective probabilities. An assumption is made so that the uncertain costs are normally distributed, where the parameters’ mean and standard deviation collected from respondents indicating percentiles at 5% and 95% for doubtful cost elements. In order to identify the project managers risk attitude, the risk premium method will be used to calculate where the expected value is produced by Pearson-Tukey method. The expected value is subtracted with certainty equivalent, which consist of the budgeted amount. The study was conducted through interviews in order to obtain the costs and the subjective probabilities. The study results show that it is difficult to determine which option is the best, or the worst, by just looking at the expected value. The risk premium shows that each option is a risk averse choice for the project managers, since the risk premium was greater than zero. / Denna studie syftar till att visa hur subjektiv osäkerhetsanalys kan tillämpas vid kostnadskalkyler samt undersöka riskattityder vid budgetering av byggprojekt. Undersökningen är en fallstudie vid Biogas Mellannorrland och fokuserar på fyra lokaliseringsalternativ för biogasanläggning. För varje alternativ görs en kostnadskalkyl i form av ett antal fasta och rörliga delkostnader, där de senare är osäkra. Som metod för att skatta subjektiva sannolikheter har Pearson-Tukey-metoden tillämpats. Ett antagande görs att osäkra kostnader är normalfördelade, där parametrarna väntevärde och standardavvikelse hämtas från respondenter som anger percentilerna vid 5 % och 95 % för osäkra delkostnader. För att identifiera projektledarnas riskattityd, kommer riskpremium att beräknas, där det förväntade värdet tas fram med Pearson-Tukey- metoden. Det förväntade värdet subtraheras med säkerhetsekvivalensen, vilket utgörs av det budgeterade beloppet. Studien har genomförts med hjälp av intervjuer för att få fram kostnader och subjektiva sannolikheter. Studiens resultat visar att det är svårt att avgöra vilka alternativ som är bäst eller sämst genom att bara titta på väntevärdet. Med riskpremium visar alla alternativ att projektledarna gör ett riskaversivt val, eftersom riskpremium blev större än noll.