Thesis (Ph. D.)--Oregon State University, 2008. / Printout. Includes bibliographical references (leaves 150-155). Also available on the World Wide Web.

Thesis (M.Sc.)(Chemistry)--University of Pretoria, 2006. / Includes summary. Includes bibliographical references. Available on the Internet via the World Wide Web.

With the upsurge of Web 2.0, customers are able to share their opinions and feelings about products and services, politics, economic shifts, current events and any number of other topics on the Web. This information, if leveraged effectively, can provide rich and valuable insights, such as: input for vendors to create successful marketing strategies, understanding of areas of improvement in products and services and tracking political opinion. The problem with this information is that it is unorganised and unstructured, therefore, it is difficult to assess automatically and in bulk. Studies in the field of sentiment analysis aim to provide a solution to determining the polarities of, and gain an overview of, the wider public opinion behind certain topics in a large volume of textual data. This research provides a novel framework and a solid, practical implementation of the proposed framework for fine-grained sentiment analysis. The framework supports mixed-opinion text and multiword expressions when analysing the sentiments expressed and the aspects that those sentiments relate to. This research uses datasets across two domains in the customer reviews area (phone products and hotel services) to evaluate the proposed framework for its reliability and validity. A sizeable performance improvement was noted whereby the proposed methodology yielded a result of 91.3% accuracy in sentiment classification, as compared to the baseline (SentiWordNet), which had a result of 71.0%. In addition, an accuracy of 92.5% was observed for the aspect analysis automatically generated across the two domains tested.

No description available.

No description available.

The test - retest reliability coefficients of the values of measurements made on the pH, the Pco₂, the standard bicarbonate, the base excess, and the buffer base, of arterial whole blood were estimated for a group of 30 male subjects with the use of the Astrup Micro-equipment, the Siggaard-Andersen revised Nomogram, and a testing program on three successive mornings. Reliability coefficients were estimated for these parameters in both the resting and the post-exercise conditions, and the intra-indivldual and the inter-individual variances were estimated for each reliability coefficient. It was found that the standard bicarbonate and the base excess had the more reliable values, and that the pH and PC0₂ had less reliable values. It was shown that the values of measurements of pH, standard bicarbonate and base excess were significantly different on Day 1 from those values on Days 2 and 3. This effect was attributed to apprehension towards the testing experience and to the strangeness of the testing environment. The measurement errors of the Astrup Micro-equipment and the Siggaard-Andersen revised Nomogram were estimated for the individual parameters. It was found that the buffer base was the only parameter in which the pH meter measurement error variance was large enough to have a decided effect upon the value of the reliability coefficient. The Nomogram measurement error variances were so small that they could be considered negligible. The temporal measurement error of the collection of blood - the error inherent in the time difference between the collection of successive tubes of blood - was investigated for the pH parameter. This temporal measurement error was shown to be practically negligible although statistically significant in the resting condition, but was shown to have both practical importance and statistical significance in the post-exercise condition. / Education, Faculty of / Curriculum and Pedagogy (EDCP), Department of / Graduate

Using data from Hedge Fund Research, Inc. (HFR), this study adapts and expands on existing methods in survival analysis in an attempt to investigate whether hedge funds mortality can be predicted on the basis of certain hedge funds characteristics. The main idea is to determine the characteristics which contribute the most to the survival and failure probabilities of hedge funds and interpret them. We establish hazard models with time-independent covariates, as well as time-varying covariates to interpret the selected hedge funds characteristics. Our results show that size, age, performance, strategy, annual audit, fund offshore and fund denomination are the characteristics that best explain hedge fund failure. We find that 1% increase in performance decreases the hazard by 3.3%, the small size and the less than 5 years old hedge funds are the most likely to die and the event-driven strategy is the best to use as compare to others. The risk of death is 0.668 times lower for funds who indicated that an annual audit is performed as compared to the funds who did not indicated that an annual audit is performed. The risk of death for the offshore hedge funds is 1.059 times higher than the non-offshore hedge funds.

Ph.D. / Harmonic time series are often used to describe the periodic nature of a time series, for example the periodic nature of a variable star’s observed light curve. Statistical methods for determining the number of harmonic components to include in harmonic time series are limited. In this thesis a stepwise bootstrap procedure based on a F-type statistic is suggested. The performance of the stepwise procedure is compared to that of Schwartz’s Bayesian Criterion (SBC) and a procedure based on a statistic described by Siegel (1980). Harmonic series with correlated noise terms and irregularly spaced observations are also considered. Tests to detect changes in harmonic parameters are also derived in this thesis. A cumulative sum statistic to test for constant amplitude is derived. It is shown that testing for constant amplitude is equivalent to testing for constant slope in simple linear regression. We also derive a likelihood ratio statistic to test for constant amplitude. It is shown that the latter likelihood ratio statistic is asymptotically equivalent to the cumulative sum statistic. These statistics are compared to a quadratic form statistic used by Koen (2009). Likelihood ratio tests are also derived for detecting changes in the frequency or phase of harmonic time series. Graphical devices to aid in diagnostic checking are suggested.

This paper aims to show how people use and understand metaphorical language in conversational discourse. Specifically, I examine how metaphorical language has the potential to be either effective or ineffective in its usage, and how they are bound to the contextual environment of the conversation. This particular setting is a conversation between a researcher and a participant involved in a therapeutic program. Metaphorical language is shown to be helpful for understanding difficult subjects; however, I found most metaphorical occurrences ineffective in meaning-making. Often these ineffective metaphors are elaborated or repeated throughout the discourse event, creating problems with cohesion and understanding. Metaphor use in conversation is an effective rhetorical tool for creating meaning, but it is also a problematic device when it comes to aligning participants' conversational goal.

Over the past decade long distance marathon running has become an important recreational activity. There is evidence that males with high levels of physical activity have some impairment of fertility. In order to investigate this further, 24 male marathon runners were studied over a period of a year. Each runner was assessed at regular intervals using hormonal profiles, anthropomorphic indices and semen evaluation. The training time and distance run increased progressively over the first five months of the study as the runners prepared for the Two Oceans marathon. Analysis of the serum hormonal profiles in this longitudinal study showed that the prolactin level increased when comparing the initial study month with the rest of the year and the progesterone level decreased. However the luteinizing hormone (LH), follicle stimulating hormone (FSH), testosterone and estradiol (E2) levels remained unchanged. When the runners were divided into a high and low training group according to the distance run in the preceding week, the only significant difference was the lower mean serum FSH level in the high training group. A decrease in semen volume was demonstrated as the training time increased. This trend was reversed as the runners' training decreased after the Two Oceans marathon. The percentage of morphologically normal spermatozoa showed an initial significant decrease in the first month of training. However, no significant difference was observed throughout the rest of year. An overall downward trend in semen motility in the first 5 months of the study was shown but this was only significant if the first and fifth study months were compared. The decrease in semen motility coincided with the period of maximum training. Since patients with an adequate sperm count but decreased motility have impaired fertility this finding is of considerable importance. In addition to the decrease in motility, there was a decrease in the percentage of morphologically normal spermatozoa when the initial month of low physical activity (December) was compared to all of the subsequent months analysed. This, too, is an important finding as the percentage of morphologically normal spermatozoa correlates directly with fertilisation and pregnancy rates. When the results were analysed in the high and low training months there was a significant difference in mean semen count and semen morphology. The mean count was higher in the high training group and this group also had a significantly higher normal morphology. However, there was no significant difference in semen volume and motility in the high and low training groups.