Ho, Siu Tung. / Thesis (M.Phil.)--Chinese University of Hong Kong, 2011. / Includes bibliographical references (p. 43-44). / Abstracts in English and Chinese. / Chapter 1 --- Introduction --- p.1 / Chapter 1.1 --- Introduction --- p.1 / Chapter 2 --- Methodology and Algorithm --- p.6 / Chapter 2.1 --- Testing one. homogeneous cluster --- p.8 / Chapter 3 --- Simulation Study --- p.17 / Chapter 3.1 --- Simulation Plan --- p.19 / Chapter 3.1.1 --- One single cluster --- p.19 / Chapter 3.1.2 --- Two separated clusters --- p.20 / Chapter 3.2 --- Measure of Performance --- p.26 / Chapter 3.3 --- Simulation Results --- p.27 / Chapter 3.3.1 --- One single cluster --- p.27 / Chapter 3.3.2 --- Two separated clusters --- p.30 / Chapter 4 --- Conclusion and further research --- p.36 / Chapter 4.1 --- Constructing Data depth --- p.38 / Bibliography --- p.43

Lithium ion batteries are ubiquitous in modern life, from powering hand-held electronic devices to electric vehicles. And with the necessary drive toward renewable energy sources like solar and wind, electricity storage for the grid promises to drive up the demand for higher performing, less expensive, safer, and more environmentally friendly secondary batteries. Recent research has theorized that replacing halogens in batteries’ electrolytes with non-halogens can yield desirable performance characteristics while eliminating the most dangerous and problematic chemicals. This thesis explores the possibility that a similar approach can be taken with the cathodes of lithium ion batteries. The active material in a cathode is a salt composed of an alkali cation – the positive lithium ion, and a negative ion – usually a metal oxide. Replacing the negative ion with a superhalogen, which is more electronegative than the most electronegative element, may yield comparable electronic properties to current cathode materials while also opening up opportunities to research materials previously not considered for lithium ion battery cathodes.

Wild bird reservoirs of influenza A contribute to the overall genetic diversity of influenza, an increased range of endemic areas, as well as, transmission methods not commonly seen in human infections. These additions to influenza transmission increase the threat posed to human populations. Therefore, understanding the patterns of transmission of influenza A subtypes in avian hosts, as well as the environmental variables associated with transmission, is paramount to creating effective surveillance programs and forecasting potential areas of high genetic changes. Using a dataset of ~151,000 birds sample for avian influenza in the US and Canada from 1986-2017, we explore spatial patterns of influenza genotypes and model the environmental niches where certain types are found. Cluster analysis and niche modeling indicate overlap but also imperfect concordance between where each subtype of avian influenza was found and where each was predicted to circulate in wild bird populations. Overall, the Midwest and New England regions indicate higher risks of influenza A in wild birds across all flu types. In addition, the urban, wetland, and water land-cover types, as well as, low levels of human population density increase the likelihood of influenza presence in the avian populations. These results indicate that influenza transmission in wild birds is heavily affected by the activities of humans as well as the general characteristics of land cover types. Together, these results allow researchers to gain a better understanding of the spatial mechanisms of the broad scale patterns associated with influenza and the areas of particular risk associated with subtypes.

This paper gives a description of three well known clustering methods, and discusses the advantages and disadvantages of each. Then, the results of these three clustering methods are compared through examining them on a specific set of data.

The focus of this thesis is the study of stellar populations of extra-galactic glob- ular clusters (GCs) by measuring spectral indices and comparing them to simple stellar population models. We present the study of GCs in the context of tracing elliptical galaxy star formation, chemical enrichment and mass assembly. In this thesis we set out to test how can be determined about a galaxy's formation history by studying the spectra of a small sample of GCs. Are the stellar population parameters of the GCs strongly linked with the formation history of the host galaxy? We present spectra and Lick index measurements for GCs associated with 3 el- liptical galaxies, NGC 1052, NGC 3379 and NGC 4649. We derive ages, metallicities and a-element abundance ratios for these GCs using the x2 minimisation approach of Proctor & Sansom (2002). The metallicities we derive are quite consistent, for old GCs, with those derived by empirical calibrations such as Brodie & Huchra (1990) and Strader & Brodie (2004). For each galaxy the GCs observed span a large range in metallicity from approximately [Z/H]=�2 to solar. We find that the majority of GCs are more than 10 Gyrs old and that we can- not distinguish any finer, age details amongst the old GC populations. However, amongst our three samples we find two age distributions contrary to our expecta- tions. From our sample of 16 GCs associated with the 1-2 Gyr old merger remnant NGC 1052, we find no young GCs. If a significant population of GCs formed during this merger we would expect those GCs to have low mass-to-light ratios and be included in our sample of bright GCs. We find 4 young GCs in our sample of 38 around NGC 4649, an old massive cluster elliptical. There are no signs of recent star formation and therefore we do not expect any GCs to have formed within the galaxy. These results seem to indicate that the GC systems of elliptical galaxies are not strongly associated with recent field star formation. We find a correlation between the α-element abundance ratio and the metallicity for all three samples. Using Thomas, Maraston & Korn (2004) models, we measure much higher α abundance ratios for low metallicity GCs than high metallicity GCs. With current data and models we are limited in both the accuracy and the detail with which we can probe this relationship. We suggest that there are some difficulties reconciling measured GC parameters with our expectations and propose some future work which could help to resolve these and other issues.

Zugl.: Hannover, Universiẗat, Diss., 2007.

<p>Nowadays, China is one of the most focused countries in the world since it has a very high potential to overwhelm the world market. As a result, we can find lots of articles and discussion about the activities of internationalization and relocation of many inter-national firms (both MNEs and SMEs) in China. Those firms moved to China in order to either serve the huge Chinese domestic market or enjoy the cheap production costs to supply the global market. However, there is one phenomenon that is not new and seldom has researches discussing about it. This is the fact that there are also a lot of Chinese entrepreneurs who moved out of China and located in other countries. In this re-search, we study about some Chinese firms which set up their sales agencies in the sales offices and distribution clusters in four European countries which are Poland, Spain, Portugal and Italy. We focus our study on the internationalization process of those Chi-nese firms, the rationales and factors influencing the location-decision and clustering and the rationales of choosing the specific countries in Europe. However, we have our latent objective, which is to stimulate more discussion and research on this interesting phenomenon since there are few researches on it.</p><p>In our research, we find that there are similar patterns of internationalization process and location-decision of Chinese firms nevertheless the differences in background and types of business they have. They have some common rationales to undertake interna-tionalization, like to avoid fierce competition in the previous market, or to find new business opportunities and so on. Most of Chinese firms are still in the initial step of in-ternationalization process, and a very common way to fulfil their internationalization is by either the Chinese owners immigrating or registering a new firm in the host countries. The help from their friends and relatives who are often earlier arrivals in the host coun-tries and the strong connection with the suppliers in China are their competitive advan-tages. However, they have encountered some common problems, such as cultural issues, like language barriers and communication problems; lack of knowledge about local market and political issues. For localization or the way they choose their locations, we find some common locational factors, like market size, degree of competition, government policy (mainly on immigration policy), connectivity, safety, transaction costs and ethnic contacts. For the last section of our study, cluster, we find that the formation of cluster are based mainly on potential benefits of cluster, such as ability to attract more customers(market size), benefits from collaboration and information sharing, availabil-ity of specialized services and supporting firms. At last, we find a lot of interesting top-ics for future research.</p>

Thesis (Ph. D.)--University of Texas at Austin, 2006. / Vita. Includes bibliographical references.

No description available.

No description available.