2020-2021 Senior Honors Theses
Permanent URI for this collectionhttps://hdl.handle.net/10657/8168
This collection contains theses produced by Class of 2021 Honors students
Browse
Browsing 2020-2021 Senior Honors Theses by Department "Electrical and Computer Engineering, Department of"
Now showing 1 - 2 of 2
- Results Per Page
- Sort Options
Item Augmented Intelligence Approach To Educational Data Mining: Student Drop Prediction(2021-05) Freeman, KeeganEducational Data Mining (EDM) and Augmented Intelligence (AUI) are two upcoming fields in the machine learning research industry. EDM refers to the use of machine learning elements in an educational format. Typically, this is in the form of utilizing educational data to better understand the learning process. Augmented Intelligence, on the other hand, is a niche of machine learning that refers to people taking a much larger role than typical in artificial intelligence projects. For example, a professional in a given field may provide better insight as to what metrics should be weighed more when considering a given prediction. In this thesis, I review the feasibility of using Augmented Intelligence in the genre of Educational Data Mining to predict the likelihood of a student dropping a course based on demographic, study habit, and student perception information recorded through a survey. Additionally, I will be testing three optimization algorithms to see which is most beneficial in the application of this research. The goal of this research is to ultimately provide instructors with a machine learning model capable of highlighting at risk students such that the instructor can provide intervention techniques in a more timely fashion.Item Sparse Deconvolution of Pulsatile Growth Hormone Secretion in Adolescents(2020-12) Genty, Jon X.Growth hormone (GH) is secreted by cells in the anterior pituitary on two time scales: discrete pulses over minutes that occur within a 24-hr pattern. Secretion reflects the balance of stimulatory and inhibitory inputs from the hypothalamus and is influenced by gonadal steroids, stress, nutrition, and sleep/wake states. We propose a novel approach for the analysis of GH data and use this approach to quantify (i) the timing, amplitude, and the number of GH pulses and (ii) GH infusion, clearance, and basal secretion (i.e., time-invariant) rates, using serum GH sampled every 10 minutes during an eight-hour sleep study in 18 adolescents. In our method, we approximate hormonal secretory events by deconvolving GH data via a two-step coordinate descent approach. The first step utilizes a sparse-recovery approach to estimate the timing and amplitude of GH secretory events. The second step estimates physiological parameters. Our method identifies the timing and amplitude of GH pulses and system parameters from experimental and simulated data, with a median R2 of 0.93, among experimental data. Recovering GH pulses and model parameters using this approach may improve the quantification of GH parameters under different physiological and pathological conditions and the design and monitoring of interventions.