In this thesis, we describe two approaches to automatically synthesize the emotional speech of a target speaker based on the hidden Markov model for his/her neutral speech. In the interpolation based method, the basic idea is the model interpolation between the neutral model of the target speaker and an emotional model selected from a candidate pool. Both the interpolation model selection and the interpolation weight computation are determined based on a model-distance measure. We propose a monophone-based Mahalanobis distance (MBMD). In the parallel model combination (PMC) based method, our basic idea is to model the mismatch between neutral model and emotional model. We train linear regression model to describe this mismatch. And then we combine the target speaker neutral model with the linear regression model. We evaluate our approach on the synthesized emotional speech of angriness, happiness, and sadness with several subjective tests. Experimental results show that the implemented system is able to synthesize speech with emotional expressiveness of the target speaker.

Automatic image annotation is the process of assigning keywords to digital images depending on the content information. In one sense, it is a mapping from the visual content information to the semantic context information. In this thesis, we propose a novel approach for automatic image annotation problem, where the annotation is formulated as a multivariate mapping from a set of independent descriptor spaces, representing a whole image, to a set of words, representing class labels. For this purpose, a hierarchical annotation architecture, named as HANOLISTIC (Hierarchical Image Annotation System Using Holistic Approach), is dened with two layers. At the rst layer, called level-0 annotator, each annotator is fed by a set of distinct descriptor, extracted from the whole image. This enables us to represent the image at each annotator by a dierent visual property of a descriptor. Since, we use the whole image, the problematic segmentation process is avoided. Training of each annotator is accomplished by a supervised learning paradigm, where each word is represented by a class label. Note that, this approach is slightly dierent then the classical training approaches, where each data has a unique label. In the proposed system, since each image has one or more annotating words, we assume that an image belongs to more than one class. The output of the level-0 annotators indicate the membership values of the words in the vocabulary, to belong an image. These membership values from each annotator is, then, aggregated at the second layer by using various rules, to obtain meta-layer annotator. The rules, employed in this study, involves summation and/or weighted summation of the output of layer-0 annotators. Finally, a set of words from the vocabulary is selected based on the ranking of the output of meta-layer. The hierarchical annotation system proposed in this thesis outperforms state of the art annotation systems based on segmental and holistic approaches. The proposed system is examined in-depth and compared to the other systems in the literature by means of using several performance criteria.

In the present work, anatase and rutile titanium dioxide (TiO2) coated lustrous mica pigments were prepared by heterogeneous nucleation method. Anatase-rutile phase transformation of the TiO2 on mica substrate was achieved by coating very thin layers of tin (IV) oxide on mica surfaces prior to TiO2 deposition. Muscovite mica, which was used in the experiments was sieved, pre-treated with sodium bicarbonate and decantated before coating process. The surface morphology of mica titania pigments and anatase-rutile phase transformation were investigated by SEM and XRD analyse, respectively. Also, microwave-assisted synthesis of copper phthalocyanine and tetracarboxamide copper phthalocyanine pigments were carried out with phthalic anhydride and trimellitic anhydride precursors, respectively. Molecular structures of these pigments were confirmed by FT-IR and UV-visible spectroscopy analyse. Furthermore, combination pigments were obtained by the process of deposition of copper phthalocyanine pigments on mica-titania pigment substrate in dimethyl formamide solvent. FT-IR analysis and XRD analyse were performed to observe the transformations in the crystal forms of copper phthalocyanines on the substrate. The surface morphologies of copper phthalocyanines on the mica titania pigments were investigated by SEM analysis. Varying amounts of copper phthalocyanines were deposited on the mica surfaces, and nitrogen elemental analysis was performed to determine the amount of copper phthalocyanines. The resulting pigments were incorporated into alkyd based resin to prepare paint samples. L*a*b* values, gloss property, and hardness of the paint samples were determined by color measuring device, gloss meter and hardness measuring device, respectively. The resulting combination pigments obtained in this study showed improved luster, hue, and color intensity. Furthermore, in literature it was reported that these pigments have very high bleed resistance. This can be attributed to large macromolecular structure of copper phthalocyanine on the surface of mica titania pigment that prevents bleeding of the pigment from the paint. Moreover, the paint samples obtained from combination pigments showed higher hardness with respect to the paint sample of the mica titania pigment.

A mixture experiment is an experiment in which the q-ingredients {xi,i=1,...,q} are nonnegative and subject to the simplex restriction sum_{i=1}^q x_i=1 on the (q-1)-dimensional probability simplex S^{q-1}. In this work, we investigate the robust D-optimal designs for mixture experiments with consideration on uncertainties in the Scheffe's linear, quadratic and cubic model without 3-way effects. The D-optimal designs for each of the Scheffe's models are used to find the robust D-optimal designs. With uncertianties on the Scheffe's linear and quadratic models, the optimal convex combination of the two model's D-optimal designs can be proved to be a robust D-optimal design. For the case of the Scheffe's linear and cubic model without 3-way effects, we have some numerical results about the robust D-optimal designs, as well as that for Scheffe's linear, quadratic and cubic model without 3-way effects. Ultimately, we discuss the efficiency of a maxmin type criterion D_r under given the robust D-optimal designs for the Scheffe's linear and quadratic models.

Abstract The research is mainly to study the marketing strategy of the experience museum activity of which as the subject is the National Museum of Marine Biology and Aquarium overnight tour. The main purpose is to research the Museum customer satisfaction reports during 2003-2007 summer periods, analyze its relevance with the turnover and explore the relationship between the marketing factors combination of ¡§products, prices, channels, promotion¡¨, and the achievements of the museum overnight tour. This research organizes the annual summertime questionnaire statistics from year 2003 to 2007, and meantime interviews with the 6 managers in charge of the activity planning, decision-making, and implementation respectively. The study is summarized as follows: 1.The questionnaire statistics indicate that the overall customer satisfaction goes in sequence "personnel services" > "activity-related process" > " hardware facility." 2. "The difference of the experience activities", "needs for the spirit satisfaction", "high customer satisfaction" and " the collaboration within the organization" are viewed as growth factors. "The effect by integrated economic environment", "old product" "the increase of the external competitors" are as recession factors. 3. From the aspect of the product, the software services (personnel services) have high satisfaction far greater than the hardware services. Even though the lack of decent hardware, the high satisfaction with the software services relatively provides the participants the profound experience, 4. In terms of the ¡§price¡¨, the museum overnight tour package can be categorized as ¡§on the basis of patronization and users relationship oriented." High price rationality is in accordance with the continuous increasing number of participants. 5. Marketing ¡§channels¡¨: High satisfaction with the application process, and personnel services can illustrate the key successful factor that the Museum itself plays in. It also goes with the increasing number of participants. 6. In view of "promotion", the results of the questionnaire and interviews both specify that "people (friends and relatives introduction)" is the most powerful experience media followed by the Internet, the electronic media, newspapers and magazines, and broadcasting.

Ranking model construction is an important topic in information retrieval. Recently, many approaches based on the idea of ¡§learning to rank¡¨ have been proposed for this task and most of them attempt to score all documents of different queries by resorting to a single function. In this thesis, we propose a novel framework of query-dependent ranking. A simple similarity measure is used to calculate similarities between queries. An individual ranking model is constructed for each training query with corresponding documents. When a new query is asked, documents retrieved for the new query are ranked according to the scores determined by a ranking model which is combined from the models of similar training queries. A mechanism for determining combining weights is also provided. Experimental results show that this query dependent ranking approach is more effective than other approaches.

As sport teams with strong brands can enjoy a loyal fan base as well as increased gate revenue, brand studies in sport have paid close attention to sport teams' brand attributes (e.g., success, star player, stadium) as key drivers to create strong brands. However, sport consumers do not only rely on the brand attributes' utility in their team consumption, but claim and anticipate positive and personally meaningful emotional experiences. Nevertheless, a number of studies on sport branding have not identified what kinds of emotions are associated with sport teams and what their impact is. As a result, the affective nature of sport team brands remains largely unexplored. This study sought to identify how emotions associated with sport team brands are structured, from the sport consumers' perspective, and to examine the relative effectiveness of the associated emotions over brand attributes on consumer behavior in the sport advertising setting. The study identified seven key dimensions of emotions associated with sport team brands: connectedness (passion, nostalgia, supportive, and connected), elation (happy, excited, pleased, proud, optimistic, and entertained), competitiveness (competitive and aggressive), surprise (amazed, surprised, and astonished), anger (annoyed, frustrated, and rage), unhappiness (suffering, sad, regret, and dejected), and worry (fearful and anxious). Connectedness, elation, and competitiveness each was positively related to sport consumer behavior while surprise, unhappiness, and worry each was negatively related to it. Interestingly, anger had a dual (positive and negative) relationship with sport consumer behavior. The study tested the relative effectiveness of emotions over cognitive brand attributes on sport consumer behavior in a sport advertising setting using a 2 (emotion: high vs. low) x 2 (cognition: high vs. low) incomplete block design. The study found that emotional advertising works better for sport teams than rational advertising whereas combination advertising works at least equally or better than rational advertising. These results suggest the primacy of emotions over cognitive brand attributes in the context of sport advertising. Given the findings and discussions, implications for sport marketing practice and future research are discussed. / text

Malaria is a disease of major public health importance, killing approximately one million people and causing about 250 million cases of fever annually. It mostly affects children under the age of five and pregnant women in many developing countries, making it a prominent issue in international health and maternal and child health. The most aggressive form of malaria is caused by the parasite Plasmodium falciparum which is responsible for 80% of infections and 90% of deaths from malaria, and is most prevalent in sub-Saharan Africa. Public Health interventions include the implementation of prevention programs, health education, and chemotherapy. The latter has experienced multiple problems in the past years whereby resistance of the parasite to the available drugs has emerged, rendering the majority of them ineffective. Furthermore, the high cost of those drugs represents a major obstacle to their dispensation in areas of the world where the affected people are often the less fortunate. The enzyme Cytochrome b5 Reductase (cb5r) and its electron acceptor Cytochrome b5 (cb5) play a role in fatty acid elongation, cholesterol biosynthesis, and cytochrome P450-mediated detoxification of xenobiotics. Therefore, these proteins are suitable as potential novel drug targets for malaria. These two proteins have been thoroughly studied in mammals but have to be characterized in microorganisms such as fungi and parasites, including Plasmodium falciparum. It is important to note that plant cb5r has been identified as a novel herbicidal target. Considering the close phylogenetic relationship between plant cb5r and Plasmodium falciparum cb5r, we conclude that these plant inhibitors may also serve as promising candidates for a new class of antimalarial drugs against the parasite. In this project, we want to obtain the biochemical and enzymatic characterization of cb5r and cb5 in order to establish whether these two proteins represent potential novel drug targets in Plasmodium falciparum malaria. This initial work may lead to the development of novel drugs which will consequently affect the field of public health with respect to drug delivery, drug resistance, and drug chemotherapy.

Support systems including base and subgrade layers play a pivotal role in manifesting and maintaining acceptable behavior and performance of continuously reinforced concrete pavement (CRCP). In Texas, especially, use of non-erodible stabilized base layers have been recommended to prevent failures of the CRCP related with pumping and erosion of the support materials. The non-erodible base materials, however, have given high initial construction cost of the rigid pavements. For this reason, it has been desired to decrease the construction cost with acceptable long-term performance of the pavement system. The primary objective of this study is to determine acceptable combination of support properties and concrete slab thickness satisfying not only adequate structural ability but also construction expense. For this purpose, field support conditions were investigated using Falling Weight Deflectometer (FWD), Dynamic Cone Penetrometer (DCP), and static plate load test in phase one. Previously developed support analysis models for rigid pavement design were examined using finite element analysis method, which model could more accurately express field support behavior. Also, effects of each support properties including base thickness, elastic modulus of base material, and subgrade k-value were mechanistically identified on composite k-value, and a method selecting optimum combinations of the support properties completing desired composite k-value was developed in phase two. Also, CRCP behavior were examined under not only diverse structural and material conditions of the support system but also the CRCP slab thickness and transverse crack spacing due to temperature and vehicle wheel loading conditions in phase three. In phase four, maximum critical stress induced in the CRCP slab was evaluated under various combinations of support conditions and slab thickness. Effects of the support properties and the slab thickness on the critical stress in the CRCP slab were mechanistically identified, and the factor with the greatest effect was verified. Moreover, regression equations were developed to estimate the maximum critical stresses for various support properties and the CRCP slab thickness under temperature and wheel loadings. In phase five, a guideline determining optimum combination of support properties and slab thickness were proposed as aspect of initial construction cost of the CRCP. / text

The aim of this study was to investigate if a mechanical recruitment process could be a useful tool for employers to avoid discrimination. National and international law protect jobseekers from discrimination during the recruiting process. Despite this individuals frequently report that they are treated unfairly when they apply for a job. In line with this research shows that some individuals do not have the same opportunities in the labour market as the rest of the population. This study focus on discrimination based on ethnicity, age, gender or disability. Today most of the hiring decisions are based on employers professional judgement. This constitute a potential risk for discrimination since the judgement can be influenced by prejudices and stereotypes. The research shows that employers tend to measure variables which are not relevant for individuals future job performance when they make decisions in selection processes. A mechanical approach for the recruitment can broadly correct for this since it is based on standardized procedures that eliminates human judgement. The data collection of candidates strictly relate to a defined job profile which only contains criterions that are of importance for the employment. Pre-specified algorithms are later on used for combining each candidates data into an overall assessment. This regulates that the same criteria is measured for all individuals. In order to analyse the material a legal dogmatic method has been combined with a legal sociology method.