Feature Engineering in Fine-Grained Image Classification

Thesis (Ph.D.)--University of Washington, 2013

Relative Efficiencies of the Burkard 7-Day, Rotorod and Burkard Personal Samplers for Poaceae and Urticaceae Pollen under Field Conditions

Introduction: In aerobiological studies it is often necessary to compare concentration data recorded with different models of sampling instrument. Sampler efficiency typically varies from device to device, and depends on the target aerosol and local atmospheric conditions. To account for these differences inter-sampler correction factors may be applied, however for many pollen samplers and pollen taxa such correction factors do not exist and cannot be derived from existing published work. Materials and methods: In this study the relative efficiencies of the Burkard 7-Day Recording Volumetric Spore Trap, the Sampling Technologies Rotorod Model 20 and the Burkard Personal Volumetric Air Sampler were evaluated for Urticaceae and Poaceae pollen under field conditions, and the influence of wind speed and relative humidity on these efficiency relationships was assessed. Data for the two pollen taxa were collected during 2010 and 2011-12 respectively. Results: The three devices were found to record significantly different concentrations for both pollen taxa, with the exception of the 7-Day and Rotorod samplers for Poaceae pollen. Under the range of conditions present during the study wind speed was found to only have a significant impact on inter-sampler relationships involving the vertically orientated Burkard Personal sampler, whilst no interaction between relative efficiency and relative humidity was observed. Conclusions: Data collected with the three models of sampler should only be compared once the appropriate correction has been made, with wind speed taken into account where appropriate.

Determination of Alternaria spp. Habitats Using 7-Day Volumetric Spore Trap, Hybrid Single Particle Lagrangian Integrated Trajectory Model and Geographic Information System

There are many species among the Alternaria genus, which hosts on economically important crops causing significant yield losses. Less attention has been paid to fungi hosting on plants constituting substantial components of pastures and meadows. Alternaria spp. spores are also recognised as important allergens. A 7-day volumetric spore trap was used to monitor the concentration of airborne fungal spores. Air samples were collected in Worcester, England (2006–2010). Days with a high spore count were then selected. The longest episode that occurred within a five year study was chosen for modelling. Two source maps presenting distribution of crops under rotation and pastures in the UK were produced. Back trajectories were calculated using the HYSPLIT model. In ArcGIS clusters of trajectories were studied in connection with source maps by including the height above ground level and the speed of the air masses. During the episode no evidence for a long distance transport from the continent of Alternaria spp. spores was detected. The overall direction of the air masses fell within the range from South-West to North. The back trajectories indicated that the most important sources of Alternaria spp. spores were located in the West Midlands of England.

Unpacking the Link Between Entrepreneurialism and Employability: an Assessment of the Relationship Between Entrepreneurial Attitudes and Likelihood of Graduate Employment in a Professional Field

Purpose This research investigates the relationship between students’ entrepreneurial attitudes and traits and their classification of employment six months after university graduation. It aims to identify what specific attitudes and traits of entrepreneurial graduates are linked to employability in a professional or managerial field. Design/Methodology The research adopts a quantitative approach to measure the entrepreneurial drive of final-year undergraduate business school students and regresses this measurement against the employment level of the same students six months after their graduation. The employment classification of each respondent was classified as ‘professional/managerial’ or ‘non-professional/non-managerial’, in line with the Standard Occupational Classification (SOC) 2010. Findings The research found that both proactive disposition and achievement motivation were statistically linked to the likelihood of graduates being employed in a professional or managerial position six months after graduation. Originality/Value This research goes beyond existing literature linking entrepreneurship to employability to quantitatively examine what specific attitudes and traits can be linked to employability in recent graduates. By identifying the aspects of entrepreneurialism that have a relationship with employability, more information is available for educators who are designing entrepreneurial education programs and allows for greater focus on aspects that may be of greatest benefit to all students.

Barry Lieberman & Friends, Maria Larionoff and Robin McCabe, May 6, 2012

Concert program for Barry Lieberman & Friends present Guest Artists, May 6, 2012

Intervention Recommendations to Minimize the 30-Day Risk-of-Readmission for Heart Failure

Thesis (Master's)--University of Washington, 2015

A fair day's wage? Perceptions of public sector pay

There is a scarcity of evidence pertaining to the general public's perception of public sector pay. Hence, in the present study, 161 women and 149 men were asked to estimate the wages 35 public sector professions should receive annually in the fictitious nation of Maldoria, based on a comparison value of an annual income of T10,000 for general practitioners. Analysis showed that only pilots were given a higher annual income than general practitioners; miners and local government workers were also provided with relatively high annual incomes. By contrast, newscasters were provided with the lowest annual income. Participants' sex did not affect these evaluations, and other demographic variables and public sector-related information of the participants were poor predictors of their evaluations. The implications of this research on public attitudes toward wage determination are discussed, and avenues for further research highlighted.

At the end of the day: Macmillan's account of the Cuban missile crisis

Pre-publicity for the final volume of Harold Macmillan’s memoirs, At the End of the Day, stressed that it would provide the British side of the Cuban missile crisis for the first time. The Churchillian model chosen, changes required by the Cabinet Office and Macmillan’s desire to rebuke those political opponents who claimed that the crisis demonstrated a lack of British influence in Washington, however ensured a focus on his personal relationship with President Kennedy. His larding the text with contemporary observations from his diaries also skewed Macmillan’s account and, in particular, underplayed the significance of British moves at the United Nations in New York to secure a credible United Nations inspection regime and a US guarantee of the inviolability of Cuba. Careful reconstruction of Macmillan’s real-time experience of the Cuban missile crisis demonstrates the limitations of his own account of this event

Testing the weak-form market efficiency and the day of the week effects of some African countries

The aims of this work are twofold. On the one hand, it aims to find evidence supporting the presence of the weak form efficiency of several emerging African stock markets by using both parametric as well as non parametric tests. The results indicate that none of the markets are characterised by random walks with the exception of the South African stock market. On the other hand, this study aims to detect the presence of the day of the week effects of these African stock markets. Results show the existence of day of the week effects, that is the typical negative Monday and Friday positive effects in several stock markets.

A beautiful day/Um dia lindo

Tradução da peça.

Statistical classification of road pavements using near field vehicle rolling noise measurements

Low noise surfaces have been increasingly considered as a viable and cost-effective alternative to acoustical barriers. However, road planners and administrators frequently lack information on the correlation between the type of road surface and the resulting noise emission profile. To address this problem, a method to identify and classify different types of road pavements was developed, whereby near field road noise is analyzed using statistical learning methods. The vehicle rolling sound signal near the tires and close to the road surface was acquired by two microphones in a special arrangement which implements the Close-Proximity method. A set of features, characterizing the properties of the road pavement, was extracted from the corresponding sound profiles. A feature selection method was used to automatically select those that are most relevant in predicting the type of pavement, while reducing the computational cost. A set of different types of road pavement segments were tested and the performance of the classifier was evaluated. Results of pavement classification performed during a road journey are presented on a map, together with geographical data. This procedure leads to a considerable improvement in the quality of road pavement noise data, thereby increasing the accuracy of road traffic noise prediction models.

Day-ahead resource scheduling in smart grids considering Vehicle-to-Grid and network constraints

Energy resource scheduling becomes increasingly important, as the use of distributed resources is intensified and massive gridable vehicle use is envisaged. The present paper proposes a methodology for dayahead energy resource scheduling for smart grids considering the intensive use of distributed generation and of gridable vehicles, usually referred as Vehicle- o-Grid (V2G). This method considers that the energy resources are managed by a Virtual Power Player (VPP) which established contracts with V2G owners. It takes into account these contracts, the user´s requirements subjected to the VPP, and several discharge price steps. Full AC power flow calculation included in the model allows taking into account network constraints. The influence of the successive day requirements on the day-ahead optimal solution is discussed and considered in the proposed model. A case study with a 33 bus distribution network and V2G is used to illustrate the good performance of the proposed method.

Data mining techniques to support the classification of MV electricity customers

This paper describes a methodology that was developed for the classification of Medium Voltage (MV) electricity customers. Starting from a sample of data bases, resulting from a monitoring campaign, Data Mining (DM) techniques are used in order to discover a set of a MV consumer typical load profile and, therefore, to extract knowledge regarding to the electric energy consumption patterns. In first stage, it was applied several hierarchical clustering algorithms and compared the clustering performance among them using adequacy measures. In second stage, a classification model was developed in order to allow classifying new consumers in one of the obtained clusters that had resulted from the previously process. Finally, the interpretation of the discovered knowledge are presented and discussed.

Demand response programs definition supported by clustering and classification techniques

The growing importance and influence of new resources connected to the power systems has caused many changes in their operation. Environmental policies and several well know advantages have been made renewable based energy resources largely disseminated. These resources, including Distributed Generation (DG), are being connected to lower voltage levels where Demand Response (DR) must be considered too. These changes increase the complexity of the system operation due to both new operational constraints and amounts of data to be processed. Virtual Power Players (VPP) are entities able to manage these resources. Addressing these issues, this paper proposes a methodology to support VPP actions when these act as a Curtailment Service Provider (CSP) that provides DR capacity to a DR program declared by the Independent System Operator (ISO) or by the VPP itself. The amount of DR capacity that the CSP can assure is determined using data mining techniques applied to a database which is obtained for a large set of operation scenarios. The paper includes a case study based on 27,000 scenarios considering a diversity of distributed resources in a 33 bus distribution network.