Filtering by
- Status: Published
As a case study, this thesis develops methods for the analysis of large amounts of data generated from a simulated ecosystem designed to understand how mammalian biomechanics interact with environmental complexity to modulate the outcomes of predator–prey interactions. These simulations investigate how other biomechanical parameters relating to the agility of animals in predator–prey pairs are better predictors of pursuit outcomes. Traditional modelling techniques such as forward, backward, and stepwise variable selection are initially used to study these data, but the number of parameters and potentially relevant interaction effects render these methods impractical. Consequently, new modelling techniques such as LASSO regularization are used and compared to the traditional techniques in terms of accuracy and computational complexity. Finally, the splitting rules and instances in the leaves of classification trees provide the basis for future simulation with an economical number of additional runs. In general, this thesis shows the increased utility of these sophisticated statistical techniques with simulated ecological data compared to the approaches traditionally used in these fields. These techniques combined with methods from industrial Design of Experiments will help ecologists extract novel insights from simulations that combine habitat complexity, population structure, and biomechanics.
This research explores the problem of the why so few of the published algorithms enter production and furthermore, fewer end up generating sustained value. The dissertation proposes a ‘Design for Deployment’ (DFD) framework to successfully build machine learning analytics so they can be deployed to generate sustained value. The framework emphasizes and elaborates the often neglected but immensely important latter steps of an analytics process: ‘Evaluation’ and ‘Deployment’. A representative evaluation framework is proposed that incorporates the temporal-shifts and dynamism of real-world scenarios. Additionally, the recommended infrastructure allows analytics projects to pivot rapidly when a particular venture does not materialize. Deployment needs and apprehensions of the industry are identified and gaps addressed through a 4-step process for sustainable deployment. Lastly, the need for analytics as a functional area (like finance and IT) is identified to maximize the return on machine-learning deployment.
The framework and process is demonstrated in semiconductor manufacturing – it is highly complex process involving hundreds of optical, electrical, chemical, mechanical, thermal, electrochemical and software processes which makes it a highly dynamic non-stationary system. Due to the 24/7 uptime requirements in manufacturing, high-reliability and fail-safe are a must. Moreover, the ever growing volumes mean that the system must be highly scalable. Lastly, due to the high cost of change, sustained value proposition is a must for any proposed changes. Hence the context is ideal to explore the issues involved. The enterprise use-cases are used to demonstrate the robustness of the framework in addressing challenges encountered in the end-to-end process of productizing machine learning analytics in dynamic read-world scenarios.
The new spatially explicit counterfactual framework considers how spatial effects impact treatment choice, treatment variation, and treatment effects. To illustrate this new methodological framework, I first replicate a classic quasi-experimental study that evaluates the effect of drinking age policy on mortality in the United States from 1970 to 1984, and further extend it with a spatial perspective. In another example, I evaluate food access dynamics in Chicago from 2007 to 2014 by implementing advanced spatial analytics that better account for the complex patterns of food access, and quasi-experimental research design to distill the impact of the Great Recession on the foodscape. Inference interpretation is sensitive to both research design framing and underlying processes that drive geographically distributed relationships. Finally, I advance a new Spatial Data Science Infrastructure to integrate and manage data in dynamic, open environments for public health systems research and decision- making. I demonstrate an infrastructure prototype in a final case study, developed in collaboration with health department officials and community organizations.
In the last two decades, fantasy sports have grown massively in popularity. Fantasy football in particular is the most popular fantasy sport in the United States. People spend hours upon hours every year building, researching, and perfecting their teams to compete with others for money or bragging rights. One problem, however, is that National Football League (NFL) players are human and will not perform the same as they did last week or last season. Because of this, there is a need to create a machine learning model to help predict when players will have a tough game or when they can perform above average. This report discusses the history and science of fantasy football, gathering large amounts of player data, manipulating the information to create more insightful data points, creating a machine learning model, and how to use this tool in a real-world situation. The initial model created significantly accurate predictions for quarterbacks and running backs but not receivers and tight ends. Improvements significantly increased the accuracy by reducing the mean average error to below one for all positions, resulting in a successful model for all four positions.
Sports analytics refers to the implementation of data science and analytics techniques within the sports industry. Several sports analysts and team managers have utilized analytical tools to boost overall team and player performance, often through the analysis of historical data. One of the most common techniques employed in sports analytics is that of data mining–the extensive practice of analyzing data in order to extract and deliver insights and findings. Data mining projects are frequently guided with the six-step Cross Industry Standard Process for Data Mining (CRISP-DM) framework. One such sport that has extensively used data science and analytics, and data mining specifically, is that of Formula One (F1). Given the sports’ reliance on technology, race engineers working for F1 constructors often develop statistical models analyzing historical race performance to derive insight of drivers’ success. For the purposes of this project, the perspective of a race engineer working for the F1 constructor McLaren was considered. As the constructor is seeking to gain a competitive advantage for the upcoming F1 season, race performance data concerning previous seasons was collected and analyzed as part of a larger data mining project utilizing the CRISP-DM framework. Statistical models, such as linear regression and random forest, were developed to predict the number of points scored by McLaren racers and the variables most strongly contributed to such scored points. The final results point to specific lap times having to be aimed for as the most important variable in determining the number of points gained, although specific locations also seem prone to McLaren race success. These results in turn will be utilized to develop race strategies for the upcoming season to ensure McLaren has high efficiency against its competitors.