Several wild-type poxviruses have emerged as potential oncolytic viruses (OVs), including orf virus (OrfV), and vaccinia virus (VV). Oncolytic VVs have been modified to include attenuating mutations that enhance their tumour selective nature, but these mutations also reduce overall viral fitness in cancer cells. Previous studies have shown that a VV (Western Reserve) with its E3L gene replaced with the E3L homologue from, OrfV (designated VV-E3LOrfV), maintained its ability to infect cells in vitro, but was attenuated compared to its parental VV in vivo. Our goal was to determine the safety and oncolytic potential VV-E3LOrfV, compared to wild type VV and other attenuated recombinants. VV-E3LOrfV, was unable to replicate to the same titers and was sensitive to IFN compared to its parental virus and other attenuated VVs in normal human fibroblast cells. The virus was also less pathogenic when administered in vivo. Viral replication, spread and cell killing, as measures of oncolytic potential in vitro, along with in vivo efficacy, were also observed.. The Parapoxvirus, OrfV has been shown to have a unique immune-stimulation profile, inducing a number of pro-inflammatory cytokines, as well as potently recruiting and activating a number of immune cells. Despite this unique profile, OrfV is limited in its ability to replicate and spread in human cancer cells. Various strategies were employed to enhance the oncolytic efficacy of wild-type OrfV. A transient transfection/infection screen was created to determine if any of the VV host-range genes (C7L, K1L, E3L or K3L) would augment OrfV oncolysis. Combination therapy, including the use of microtubule targeting agents, Viral Sensitizer (VSe) compounds and the addition of soluble VV B18R gene product were employed to see if they also enhance OrfV efficacy. Unfortunately, none of the strategies mentioned were able to enhance OrfV.

Access technologies often deal with the classification of several physiological and biomechanical signals. In most previous studies involving access technologies, a single classifier has been trained. Despite reported success of these single classifiers, classification accuracies are often below clinically viable levels. One approach to improve upon the performance of these classifiers is to utilize the state of- the-art multiple classifier systems (MCS). Because MCS invoke more than one classifier, more information can be exploited from the signals, potentially leading to higher classification performance than that achievable with single classifiers. Moreover, by decreasing the feature space dimensionality of each classifier, the speed of the system can be increased. MCSs may combine classifiers on three levels: abstract, rank, or measurement level. Among them, abstract-level MCSs have been the most widely applied in the literature given the flexibility of the abstract level output, i.e., class labels may be derived from any type of classifier and outputs from multiple classifiers, each designed within a different context, can be easily combined. In this thesis, we develop two new abstract-level MCSs based on "reputation" values of individual classifiers: the static reputation-based algorithm (SRB) and the dynamic reputation-based algorithm (DRB). In SRB, each individual classifier is applied to a “validation set”, which is disjoint from training and test sets, to estimate its reputation value. Then, each individual classifier is assigned a weight proportional to its reputation value. Finally, the total decision of the classification system is computed using Bayes rule. We have applied this method to the problem of dysphagia detection in adults with neurogenic swallowing difficulties. The aim was to discriminate between safe and unsafe swallows. The weighted classification accuracy exceeded 85% and, because of its high sensitivity, the SRB approach was deemed suitable for screening purposes. In the next step of this dissertation, I analyzed the SRB algorithm mathematically and examined its asymptotic behavior. Specifically, I contrasted the SRB performance against that of majority voting, the benchmark abstract-level MCS, in the presence of different types of noise. In the second phase of this thesis, I exploited the idea of the Dirichlet reputation system to develop a new MCS method, the dynamic reputation-based algorithm, which is suitable for the classification of non-stationary signals. In this method, the reputation of each classifier is updated dynamically whenever a new sample is classified. At any point in time, a classifier’s reputation reflects the classifier’s performance on both the validation and the test sets. Therefore, the effect of random high-performance of weak classifiers is appropriately moderated and likewise, the effect of a poorly performing individual classifier is mitigated as its reputation value, and hence overall influence on the final decision is diminished. We applied DRB to the challenging problem of discerning physiological responses from nonverbal youth with severe disabilities. The promising experimental results encourage further development of reputation-based multi-classifier systems in the domain of access technology research.

Access technologies often deal with the classification of several physiological and biomechanical signals. In most previous studies involving access technologies, a single classifier has been trained. Despite reported success of these single classifiers, classification accuracies are often below clinically viable levels. One approach to improve upon the performance of these classifiers is to utilize the state of- the-art multiple classifier systems (MCS). Because MCS invoke more than one classifier, more information can be exploited from the signals, potentially leading to higher classification performance than that achievable with single classifiers. Moreover, by decreasing the feature space dimensionality of each classifier, the speed of the system can be increased. MCSs may combine classifiers on three levels: abstract, rank, or measurement level. Among them, abstract-level MCSs have been the most widely applied in the literature given the flexibility of the abstract level output, i.e., class labels may be derived from any type of classifier and outputs from multiple classifiers, each designed within a different context, can be easily combined. In this thesis, we develop two new abstract-level MCSs based on "reputation" values of individual classifiers: the static reputation-based algorithm (SRB) and the dynamic reputation-based algorithm (DRB). In SRB, each individual classifier is applied to a “validation set”, which is disjoint from training and test sets, to estimate its reputation value. Then, each individual classifier is assigned a weight proportional to its reputation value. Finally, the total decision of the classification system is computed using Bayes rule. We have applied this method to the problem of dysphagia detection in adults with neurogenic swallowing difficulties. The aim was to discriminate between safe and unsafe swallows. The weighted classification accuracy exceeded 85% and, because of its high sensitivity, the SRB approach was deemed suitable for screening purposes. In the next step of this dissertation, I analyzed the SRB algorithm mathematically and examined its asymptotic behavior. Specifically, I contrasted the SRB performance against that of majority voting, the benchmark abstract-level MCS, in the presence of different types of noise. In the second phase of this thesis, I exploited the idea of the Dirichlet reputation system to develop a new MCS method, the dynamic reputation-based algorithm, which is suitable for the classification of non-stationary signals. In this method, the reputation of each classifier is updated dynamically whenever a new sample is classified. At any point in time, a classifier’s reputation reflects the classifier’s performance on both the validation and the test sets. Therefore, the effect of random high-performance of weak classifiers is appropriately moderated and likewise, the effect of a poorly performing individual classifier is mitigated as its reputation value, and hence overall influence on the final decision is diminished. We applied DRB to the challenging problem of discerning physiological responses from nonverbal youth with severe disabilities. The promising experimental results encourage further development of reputation-based multi-classifier systems in the domain of access technology research.

As multidrug resistance in Gram-negative bacilli increases, the old antibiotic colistin has rapidly gained attention as one of few last line treatment options in the form of colistin methanesulfonate (CMS), which is hydrolyzed to colistin both in vitro and in vivo. There is a dearth of knowledge on fundamental aspects of colistin, including pharmacokinetics and optimal dosing regimens. The aim of this thesis was to improve the basis for optimal colistin therapy. To be able to study colistin, an LC-MS/MS assay method was developed which is sensitive, specific and useful in both in vivo and in vitro studies. Using this method we detected a significant loss of colistin during standard laboratory procedures. This loss was characterized and quantified, the hypothesis being that the loss is mainly caused by adsorption to labware. The pharmacokinetics of colistin was studied in two populations of critically ill patients, one with normal renal function and one with renal replacement therapy. Plasma concentrations were assayed with the method above, and population modeling was employed to describe the data. The results include a previously unseen, long elimination half-life of colistin. The data from the population on renal replacement therapy was described without modeling, and showed that both CMS and colistin are cleared by hemodiafiltration. Combination therapy is an approach that is often used when treating patients infected with multidrug-resistant pathogens. The thesis discusses how the joint effect of antibiotics can be measured using colistin and meropenem as a model, and proposes a method for testing antibiotic combinations. Furthermore, a PKPD model was adapted to describe the pharmacodynamics of the combination. In conclusion, a specific and sensitive method for analysis of colistin was developed and the adsorption of colistin to materials was described. The assay method has been well accepted internationally. The pharmacokinetics of colistin and CMS was described in two important patient populations, partly with surprising results that have influenced dosages of colistin worldwide. The pharmacodynamics of combination therapy was investigated and quantified, and the methods applied could be further developed into clinically useful tools for selection of antibiotic combinations.

To solve the challenging pattern classification problem, machine learning researchers have extensively studied Multiple Classifier Systems (MCSs). The motivations for combining classifiers are found in the literature from the statistical, computational and representational perspectives. Although the results of classifier combination does not always outperform the best individual classifier in the ensemble, empirical studies have demonstrated its superiority for various applications. A number of viable methods to design MCSs have been developed including bagging, adaboost, rotation forest, and random subspace. They have been successfully applied to solve various tasks. Currently, most of the research is being conducted on the behavior patterns of the base classifiers in the ensemble. However, a discussion from the learning point of view may provide insights into the robust design of MCSs. In this thesis, Generalized Exhaustive Search and Aggregation (GESA) method is developed for this objective. Robust performance is achieved using GESA by dynamically adjusting the trade-off between fitting the training data adequately and preventing the overfitting problem. Besides its learning algorithm, GESA is also distinguished from traditional designs by its architecture and level of decision-making. GESA generates a collection of ensembles and dynamically selects the most appropriate ensemble for decision-making at the local level. Although GESA provides a good improvement over traditional approaches, it is not very data-adaptive. A data- adaptive design of MCSs demands that the system can adaptively select representations and classifiers to generate effective decisions for aggregation. Another weakness of GESA is its high computation cost which prevents it from being scaled to large ensembles. Generalized Adaptive Ensemble Generation and Aggregation (GAEGA) is an extension of GESA to overcome these two difficulties. GAEGA employs a greedy algorithm to adaptively select the most effective representations and classifiers while excluding the noise ones as much as possible. Consequently, GAEGA can generate fewer ensembles and significantly reduce the computation cost. Bootstrapped Adaptive Ensemble Generation and Aggregation (BAEGA) is another extension of GESA, which is similar with GAEGA in the ensemble generation and decision aggregation. BAEGA adopts a different data manipulation strategy to improve the diversity of the generated ensembles and utilize the information in the data more effectively. As a specific application, the classification of time series data is chosen for the research reported in this thesis. This type of data contains dynamic information and proves to be more complex than others. Multiple Input Representation-Adaptive Ensemble Generation and Aggregation (MIR-AEGA) is derived from GAEGA for the classification of time series data. MIR-AEGA involves some novel representation methods that proved to be effective for time series data. All the proposed methods including GESA, GAEGA, MIR-AEGA, and BAEGA are tested on simulated and benchmark data sets from popular data repositories. The experimental results confirm that the newly developed methods are effective and efficient.

Abstract   Title:                               Goodwill – differences and similarities between how IFRS and U.S. GAAP treats goodwill   Level:                              One year master, 15 credits   Author:                           Oscar Larsen and Thomas Karlsson   Supervisor:                     Leif Carlsson   Examiner:                      Cecilia Lindh   Year of publication:      2011   The main issue:              What are the differences and similarities of the treatment of goodwill between the U.S. GAAP and IFRS? What can the transition from U.S. GAAP to IFRS imply to companies regarding goodwill?   Aim:                                The aim with this thesis is to treat differences and similarities between U.S. GAAP and IFRS regarding how goodwill is generated and distributed in a business combination and also how the process of an impairment test of goodwill is carried through. The aim is further to treat possible effects that a transition may imply to companies regarding goodwill.   Method:                          In this study a qualitative research method has been used where the gathered information has been collected from interviews that has been analyzed and compared with the theoretical studies.   Conclusion:                    Differences described in theory doesn’t have to be confirmed with practices though it depends on how different companies in different industries interpret and apply these regulations   Keywords:                      Goodwill, IFRS, U.S. GAAP, impairment, business combination

To solve the challenging pattern classification problem, machine learning researchers have extensively studied Multiple Classifier Systems (MCSs). The motivations for combining classifiers are found in the literature from the statistical, computational and representational perspectives. Although the results of classifier combination does not always outperform the best individual classifier in the ensemble, empirical studies have demonstrated its superiority for various applications. A number of viable methods to design MCSs have been developed including bagging, adaboost, rotation forest, and random subspace. They have been successfully applied to solve various tasks. Currently, most of the research is being conducted on the behavior patterns of the base classifiers in the ensemble. However, a discussion from the learning point of view may provide insights into the robust design of MCSs. In this thesis, Generalized Exhaustive Search and Aggregation (GESA) method is developed for this objective. Robust performance is achieved using GESA by dynamically adjusting the trade-off between fitting the training data adequately and preventing the overfitting problem. Besides its learning algorithm, GESA is also distinguished from traditional designs by its architecture and level of decision-making. GESA generates a collection of ensembles and dynamically selects the most appropriate ensemble for decision-making at the local level. Although GESA provides a good improvement over traditional approaches, it is not very data-adaptive. A data- adaptive design of MCSs demands that the system can adaptively select representations and classifiers to generate effective decisions for aggregation. Another weakness of GESA is its high computation cost which prevents it from being scaled to large ensembles. Generalized Adaptive Ensemble Generation and Aggregation (GAEGA) is an extension of GESA to overcome these two difficulties. GAEGA employs a greedy algorithm to adaptively select the most effective representations and classifiers while excluding the noise ones as much as possible. Consequently, GAEGA can generate fewer ensembles and significantly reduce the computation cost. Bootstrapped Adaptive Ensemble Generation and Aggregation (BAEGA) is another extension of GESA, which is similar with GAEGA in the ensemble generation and decision aggregation. BAEGA adopts a different data manipulation strategy to improve the diversity of the generated ensembles and utilize the information in the data more effectively. As a specific application, the classification of time series data is chosen for the research reported in this thesis. This type of data contains dynamic information and proves to be more complex than others. Multiple Input Representation-Adaptive Ensemble Generation and Aggregation (MIR-AEGA) is derived from GAEGA for the classification of time series data. MIR-AEGA involves some novel representation methods that proved to be effective for time series data. All the proposed methods including GESA, GAEGA, MIR-AEGA, and BAEGA are tested on simulated and benchmark data sets from popular data repositories. The experimental results confirm that the newly developed methods are effective and efficient.

The growing need for designing and implementing reliable voice-based human-machine interfaces has inspired intensive research work in the field of voice-enabled systems, and greater robustness and reliability are being sought for those systems. Speech recognition has become ubiquitous. Automated call centers, smart phones, dictation and transcription software are among the many systems currently being designed and involving speech recognition. The need for highly accurate and optimized recognizers has never been more crucial. The research community is very actively involved in developing powerful techniques to combine the existing feature extraction methods for a better and more reliable information capture from the analog signal, as well as enhancing the language and acoustic modeling procedures to better adapt for unseen or distorted speech signal patterns. Most researchers agree that one of the most promising approaches for the problem of reducing the Word Error Rate (WER) in large vocabulary speech transcription, is to combine two or more speech recognizers and then generate a new output, in the expectation that it provides a lower error rate. The research work proposed here aims at enhancing and boosting even further the performance of the well-known Recognizer Output Voting Error Reduction (ROVER) combination technique. This is done through its integration with an error filtering approach. The proposed system is referred to as cROVER, for context-augmented ROVER. The principal idea is to flag erroneous words following the combination of the word transition networks through a scanning process at each slot of the resulting network. This step aims at eliminating some transcription errors and thus facilitating the voting process within ROVER. The error detection technique consists of spotting semantic outliers in a given decoder's transcription output. Due to the fact that most error detection techniques suffer from a high false positive rate, we propose to combine the error filtering techniques to compensate for the poor performance of each of the individual error classifiers. Experimental results, have shown that the proposed cROVER approach is able to reduce the relative WER by almost 10% through adequate combination of speech decoders. The approaches proposed here are generic enough to be used by any number of speech decoders and with any type of error filtering technique. A novel voting mechanism has also been proposed. The new confidence-based voting scheme has been inspired from the cROVER approach. The main idea consists of using the confidence scores collected from the contextual analysis, during the scoring of each word in the transition network. The new voting scheme outperformed ROVER's original voting, by up to 16% in terms of relative WER reduction.

Taiwan administration district division and politic landscape began to appear significant combination and revolution since the issuance of the amendment of¡§ Law of Local Politics¡¨ on Feb. 3, 2010. The Kaohsiung county and Kaohsiung city has been combined to a totally new political structure which will have a significant impact on the reorganization of township governments. Therefore the issues such as the governance interrelationship among local government organizations and their own governance relationship, the combination method of township regions, and the revolution of organization structure are important and deserve for further investigation. The study takes the Hunei Township of Kaohsiung County as an example. The township local leaders, focus groups and other important interested parties are interviewed. Through the deep interview method and the SWOT analysis, this study constructs the problems and investigates possible solutions related to the development of township government after the combination of the county and the city, in order to reduce the obstruction of organization reform. Therefore, the main aims of the study are the followings: (1) discuss the rationality and justness of an administrative area combination; (2) discuss the supporting measures that are necessary for increasing organizational effectiveness after the combination; (3) discuss the local fiscal division issues after the combination; and (4) analyze simplified personnel and taxation efficiency. The study aims at providing suggestions for the future sustainable development of Hunei township government after the combination of Kaohsiung district, which will be helpful for the development of other villages and towns.

This paper deals with the applications of symbolic computation of Mathematica 7.0 (Wolfram, 2008) in distribution theory. The purpose of this study is twofold. Firstly, we will implement some functions to extend Mathematica capabilities to handle symbolic computations of the characteristic function for linear combination of independent univariate random variables. These functions utilizes pattern-matching codes that enhance Mathematica's ability to simplify expressions involving the product and summation of algebraic terms. Secondly, characteristic function can be classified into commonly used distributions, including six discrete distributions and seven continuous distributions, via the pattern-matching feature of Mathematica. Finally, several examples will be presented. The examples include calculating limit of characteristic function of linear combinations of independent random variables, and applications of coded functions and illustrate the central limit theorem, the law of large numbers and properties of some distributions.