No description available.

No description available.

No description available.

In psychiatric settings, the use of seclusion and/or restraints can be emotionally and psychologically traumatizing for patients. Patients often experience these interventions as inhumane and humiliating, and such interventions can have physical and mental adverse effects and in some cases can be fatal. This study examined the role of demographic, clinical, and hospital variables in predicting seclusion and/or restraint episodes in adult psychiatric inpatients. A total of 395 patients were included in the study. Adult psychiatric inpatients previously restrained (n = 91) were compared to psychiatric inpatients never restrained (n = 304). A binary logistic regression research design was used to examine the relationship of demographic variables, clinical variables, and hospital variables on the likelihood of being placed in seclusion or restraints. The results yielded age as a significant predictor for patients being restrained. Also, individuals diagnosed with bipolar disorder were less likely to experience a seclusion and/or restraint event than patients diagnosed with depressive disorder or within the schizophrenia spectrum. In addition, findings suggest that adult psychiatric inpatients that experienced restraint episodes were restrained within the 1st month of admission, during the weekday and during the 1st shift. In summary, given the findings from this study, knowledge of risk factors that precede patient restraint could enhance education and provide staff with information necessary to meet the clinical needs of the psychiatric inpatient population. Research indicates that the use of seclusion and restraint has decreased followed by implementation of educational programs designed to help staff assess patient clinical care needs and develop more therapeutically appropriate alternatives (Bower et al., 2003). By being aware of possible risk factors associated with seclusion and/or restraint, mental health providers can use early intervention and prevention strategies to reduce the use of seclusion and/or restraint. This would provide safer environments for mental health patients receiving treatment.

Low volume high value (LVHV) supply chains such as airline manufacturing, power plant construction, and shipbuilding are especially susceptible to risks. These industries are characterized by long lead times and a limited number of suppliers that have both the technical know-how and manufacturing capabilities to deliver the requisite goods and services. Disruptions within the supply chain are common and can cause significant and costly delays. Although supply chain risk management and supply chain reliability are topics that have been studied extensively, most research in these areas focus on high vol- ume supply chains and few studies proactively identify risks. In this research, we develop methodologies to proactively and quantitatively identify and mitigate supply chain risks within LVHV supply chains. First, we propose a framework to model the supply chain system using fault-tree analysis based on the bill of material of the product being sourced. Next, we put forward a set of mathematical optimization models to proactively identify, mitigate, and resource at-risk suppliers in a LVHV supply chain with consideration for a firm’s budgetary constraints. Lastly, we propose a machine learning methodology to quan- tify the risk of an individual procurement using multiple logistic regression and industry available data, which can be used as the primary input to the fault tree when analyzing overall supply chain system risk. Altogether, the novel approaches proposed within this dissertation provide a set of tools for industry practitioners to predict supply chain risks, optimally choose which risks to mitigate, and make better informed decisions with respect to supplier selection and risk mitigation while avoiding costly delays due to disruptions in LVHV supply chains.

E-handel har varit en snabbt växande sektor de senaste åren och förväntas fortsätta växa i samma takt under de närmsta. Detta har öppnat upp nya möjligheter för företag som försöker sälja sina produkter och tjänster, men det tvingar dem även att utnyttja dessa möjligheter för att vara konkurrenskraftiga. En intressant möjlighet som vi har valt att fokusera detta arbete på är förmågan att använda kunddata, som inte varit tillgänglig i fysiska butiker, till att identifiera mönster i kundbeteenden. Förhoppningsvis ger detta en ökad förståelse för kunderna och gör det möjligt att förutspå framtida beteenden. Vi fokuserade specifikt på att skilja mellan potentiella köpare och faktiska köpare, med avsikt att identifiera nyckelfaktorer som avgör ifall en kund genomför ett köp eller ej. Detta gjorde vi genom att använda Binary Logistic Regression, en algoritm som använder övervakad maskininlärning för att klassificera en observation mellan två klasser. Vi lyckades ta fram en modell som förutsåg om en kund skulle genomföra ett köp eller ej med en noggrannhet på 88%. / E-commerce has been a rapidly growing sector during the last years, and are predicted to continue to grow as fast during the next ones. This has opened up a lot of opportunities for companies trying to sell their products or services, but it is also forcing them to exploit these opportunities before their competitors in order to not fall behind. One interesting opportunity we have chosen to focus this thesis on is the ability to use customer data, that has not been available with physical stores, to identify customer behaviour patterns and develop a better understanding for the customers. Hopefully this makes it possible to predict customer behaviour. We specifically focused on distinguishing possible-buyers from buyers, with the intent of identifying key factors that affect whether the customer performs a purchase or not. We did this using Binary Logistic Regression, a supervised machine learning algorithm that is trained to classify an input observation. We managed to create a model that predicted whether or not a customer was a possible-buyer or buyer with an accuracy of 88%.

The population trajectory of woodland caribou (Rangifer tarandus caribou) in Newfoundland is currently determined by low calf survival due to high predation rates during the first 6-8 weeks after parturition. Most caribou in Newfoundland congregate and give birth in open calving grounds; consequently, in order to investigate predator-prey interactions, design research, and develop mitigation strategies, the geographic extent of the caribou calving grounds must be properly identified. We used VHF telemetry locations of caribou calves, collected from 2003-2010, to determine the spatial and temporal extent of caribou calving grounds in three study areas in Newfoundland. We put GPS collars on 47 black bears (Ursus americanus) in 3 caribou ranges where bears are having a significant impact on caribou recruitment by preying on calves during the calving season. Bear movements were greatest during the calving season, potentially increasing encounters with calves. Some bears migrated to the calving grounds just prior to caribou parturition, indicating deliberate broad-scale selection of areas of high calf density. Bears displayed interannual fidelity to calving ground usage patterns during the calving season, with some bears using the calving grounds every year, while others did not. We estimated the probability of a bear spending time in the calving grounds during the calving season as a function of the bear’s sex and mean distance to the calving grounds with logistic regression. We found that as distance increased, the odds of a bear spending time in the calving grounds decreased, and that at any given distance the odds were greater for male bears than for female bears. Our results indicate that some bears in Newfoundland are likely caribou calf predators, while others are not, and that the sex and broad-scale distribution of bears influenced the probability of a bear participating in calf predation during the calving season. The probability distribution of calf-visiting bears could be used to develop management practices to mitigate the impact of bear predation on declining caribou herds in Newfoundland.

Background. Due to an increasing number of credit card defaulters, companies arenow taking greater precautions when approving credit applications. When a customermeets certain requirements, credit card firms typically use their experience todecide whether to grant them a credit card. Additionally, a few machine learningmethods have been applied to support the final decision. Objectives. The aim of this thesis is to compare the accuracy of logistic regressionclassifier, random forest classifier, and support vector classifier with the ensemblebagging classifier for predicting credit card approval. Methods. This thesis follows a method called general experimentation to determinethe most accurate classification technique for predicting credit card approval. Thedataset is taken from Kaggle, which contains information about credit card applications.The selected algorithms are trained with training data and validate themusing validation data then evaluate their performance on the testing data by usingmetrics such as accuracy, precision, recall, F1 score, and ROC curve. Now ensemblelearning bagging technique is applied to combine the predictions of these multiplemodels using majority voting to create an ensemble model. Finally, the performanceof the ensemble model was evaluated on the testing data and compared its accuracyto that of the individual models to identify the most accurate classification techniquefor predicting credit card approval. Results. Among the four selected machine learning algorithms, the random forestclassifier performed better with an accuracy of 88.41% on the testing dataset.The second-best algorithm is the ensemble bagging classifier, with an accuracy of84.78%. Hence, the random forest classifier is the most accurate algorithm for predictingcredit card approval. Conclusions. After evaluating various classifiers, including logistic regression classifier,random forest classifier, support vector classifier, and ensemble bagging, it wasobserved that the random forest classifier outperformed the other models in termsof predicting accuracy. This indicates that the random forest classifier was better atpredicting credit card approval.

No description available.

No description available.