Search Content

Data Management Behind Machine Learning

Description

This thesis dives into the world of artificial intelligence by exploring the functionality of a single layer artificial neural network through a simple housing price classification example while simultaneously considering its impact from a data management perspective on both the software and hardware level. To begin this study, the universally…

This thesis dives into the world of artificial intelligence by exploring the functionality of a single layer artificial neural network through a simple housing price classification example while simultaneously considering its impact from a data management perspective on both the software and hardware level. To begin this study, the universally accepted model of an artificial neuron is broken down into its key components and then analyzed for functionality by relating back to its biological counterpart. The role of a neuron is then described in the context of a neural network, with equal emphasis placed on how it individually undergoes training and then for an entire network. Using the technique of supervised learning, the neural network is trained with three main factors for housing price classification, including its total number of rooms, bathrooms, and square footage. Once trained with most of the generated data set, it is tested for accuracy by introducing the remainder of the data-set and observing how closely its computed output for each set of inputs compares to the target value. From a programming perspective, the artificial neuron is implemented in C so that it would be more closely tied to the operating system and therefore make the collected profiler data more precise during the program's execution. The program is designed to break down each stage of the neuron's training process into distinct functions. In addition to utilizing more functional code, the struct data type is used as the underlying data structure for this project to not only represent the neuron but for implementing the neuron's training and test data. Once fully trained, the neuron's test results are then graphed to visually depict how well the neuron learned from its sample training set. Finally, the profiler data is analyzed to describe how the program operated from a data management perspective on the software and hardware level.

ContributorsRichards, Nicholas Giovanni (Author) / Miller, Phillip (Thesis director) / Meuth, Ryan (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)

Created2018-05

Medium-scale Automation System Design and Implementation

Description

Under the supervision of Pofessor Robert Hammond, I handled the programming and record-keeping needs of a project at the Arizona Public Service Solar Test and Research Center (STAR). In the course of the first year that I worked there, I became aware that STAR's Data Management System (DMS) was in…

Under the supervision of Pofessor Robert Hammond, I handled the programming and record-keeping needs of a project at the Arizona Public Service Solar Test and Research Center (STAR). In the course of the first year that I worked there, I became aware that STAR's Data Management System (DMS) was in need of an overhaul due to an increasingly volatile date set that was quickly growing in size. STAR management was looking for a software system that would retrieve and store data automatically, that would contain a friendly user-interface, that minimized space usage on a crowded hard drive, that provided quick access to charts, and that generated statistical analysis of solar plant operation. STAR's current DMS consists of four top-level procedures. The latest version of STAR's DMS began operation two and a half years ago. The goal of the following chapters is to document and critique the software development process that I used to bring the Visual Basic for Excel version of the current software components into existence. In addition, the conclusion will include a look into the future of STAR's DMS as management introduces an Access database version for the implementation of the DMS.

ContributorsEveringham, Spencer (Author) / Pheanis, David (Thesis director) / Hammond, Robert (Committee member) / Broberg, Gregory (Committee member) / Barrett, The Honors College (Contributor)

Created2003-05

Data management planning before supercomputing and A collaborative approach in research data sharing at ASU

Description

A two-part presentation from the ASU Library and Knowledge Enterprise Research Data Management Office. Presented at the 2023 Rocky Mountain Advanced Computing Consortium (RMACC).

Session 1: Data management planning is an integral step in the research data life cycle. Large amounts of data and lengthy code accompanying supercomputing runs are no…

A two-part presentation from the ASU Library and Knowledge Enterprise Research Data Management Office. Presented at the 2023 Rocky Mountain Advanced Computing Consortium (RMACC).

Session 1: Data management planning is an integral step in the research data life cycle. Large amounts of data and lengthy code accompanying supercomputing runs are no exception. Planning before analysis will benefit research and the researcher by providing a clear strategy for collecting, storing, analyzing, and sharing the data at the end of the research cycle. Supercomputing can require significant storage beyond scratch space, but researchers typically need to be informed of what tools are appropriate and available. Framed within the planning phase of the life cycle, this presentation presents ASU’s Storage Selector as a quick and easy tool to find the most appropriate storage resources provided by the university to help researchers choose a proper storage and management solution for their research data at the right time in their project. We will also explore the DMP Tool, developed by the California Digital Library, which provides a resource-rich platform for writing data management plans, including institutional-specific guidance, feedback request, and public plans that can be used as guides.

Session 2: This presentation overviews the ongoing working relationship between the ASU Library Open Science and Scholarly Communication division, Research Data Management Office, and Research Computing. We will explore these teams’ interdisciplinary relationships and interdependence as the institution increasingly supports open science practices and initiatives. We will include case studies regarding the decision-making process, data-sharing decisions, and opportunities and challenges when transferring research data from a high-performance computing environment to the ASU Research Data Repository. Finally, we will share lessons learned as we intentionally shepherd research data from active project management and storage to final publication and preservation.

ContributorsHarp, Matthew (Author) / Claypool, Kathryn (Author)

Created2023-05-17

Data Management and Technology Avoidance in a Free Clinic

Description

Capturing and presenting high-quality data can be challenging for free clinics due to lack of resources and technology avoidance. If free clinics are unable to present impactful data to current and potential donors, this may limit funding and restrict care provided to underserved and vulnerable populations. The following is a…

Capturing and presenting high-quality data can be challenging for free clinics due to lack of resources and technology avoidance. If free clinics are unable to present impactful data to current and potential donors, this may limit funding and restrict care provided to underserved and vulnerable populations. The following is a quality improvement project which addresses utilization of information systems within a free clinic. For one month, volunteer providers completed appointment summary forms for each patient seen in the clinic. Electronic form submissions (E=110) were compared to paper form submissions (P=196), with quality of data determined by form completeness scores. Welch’s t-test was used to determine statistical significance between electronic and paper form completeness scores (E=9.7, P=8.5) (p < .001). Findings suggest that utilization of electronic data collection tools within a free clinic produce more complete and accurate data. Barriers associated with technology utilization in this under-resourced environment were substantial. Findings related to overcoming some of these barriers may be useful for future exploration of health information technology utilization in under-resourced and technology avoidant settings. Results warrant future investigation of the relationship between quality of free clinic data, information management systems, provider willingness to utilize technology and funding opportunities in free clinics.

ContributorsCorcoran, Abigail T. (Author) / Thrall, Charlotte (Thesis advisor) / Arizona State University. College of Nursing & Healthcare Innovation (Contributor)

Created2021-04-24

MICROSOFT POWER BI CASE STUDY: AN EFFICIENT BUSINESS INTELLIGENCE SOLUTION

Description

This thesis investigates the use of MS Power BI in the case company’s heterogeneous computing environment. The empirical evidence was collected through the authors’ own observations and exposure to the modeling of dashboards, other supported external findings from interviews, published articles, academic journals, and speaking with leading experts at the…

This thesis investigates the use of MS Power BI in the case company’s heterogeneous computing environment. The empirical evidence was collected through the authors’ own observations and exposure to the modeling of dashboards, other supported external findings from interviews, published articles, academic journals, and speaking with leading experts at the WA ‘Dynamic Talks Seattle/Redmond: Big Data Analytics’ conference. Power BI modeling is effective for advancing the development of statistical thinking and data retrieving skills, finding trends and patterns in data representations, and making predictions. Computer-based data modeling gave meaning to math results, and supported examining implications of these results with simple charts to improve perception. Querying and other add-ins that would be seen as affordances when using other BI softwares, with some complexity removed in Power BI, make modeling data an easier undertaking for report builders. Using computer-based qualitative data analysis software, this paper details opportunities and challenges of data modeling with dashboards. Simple linear regression is used for case study use only.

ContributorsKusen, Alexandra Jeshua (Co-author) / Briones, Jared (Co-author) / Fugleberg, Aaron (Co-author) / Lin, Amy (Co-author) / Simonson, Mark (Thesis director) / Hertzel, Michael (Committee member) / Department of Finance (Contributor) / Barrett, The Honors College (Contributor)

Created2020-05