Physical Sciences and Mathematics Commons^™

Sign-A-Mander: A Mobile App That Enhances Asl Learning With Computer Vision, Sandrine Adap

Honors Theses

Several machine learning researchers have developed algorithms recognizing American Sign Language (ASL), but few have applied the algorithms to real-world situations, such as with portable ASL learning applications. This project develops a beta version of a mobile application designed to allow beginner ASL learners to practice basic ASL vocabulary and receive feedback about their signing accuracy. Building on Dongxu Li et al.’s I3D sign language recognition algorithm and 2000-word dataset, the app seeks to determine whether the I3D algorithm can sufficiently recognize a user’s motions when recorded from a mobile device and accurately classify whether or not the user signed …

Andrews University Pre-Professional Students Preparedness For A Future With Artificial Intelligence, Zachary Alignay

Honors Theses

Artificial Intelligence technology has advanced considerably over the past four years. With such rapid technological development, the question has to be asked if students are adequately educated on the implications and abilities of artificial intelligence. Are Andrews University pre-professional students prepared for future careers with artificial intelligence? To approach this question, a survey of students across multiple perspectives was conducted to sample if there was a consensus, or lack thereof, on the perception of ethics regarding artificial intelligence, to ask students how using artificial intelligence has changed their education, what purposes it can be used or cannot be used, personal …

Variable Autoencoders For Biosensor Data Augmentation, Solomon Kim

Honors Theses

Over the past decade machine learning and artificial intelligence's resurgence spawned the desire to mimic human creative ability. Initially attempts to create images, music, and text flooded the community, though little has been learned regarding constrained, one-dimensional data generation. This paper demonstrates a variational autoencoder approach to this problem. By modeling biosensor current and concentration data we aim to augment the existing dataset. In training a multi-layer neural network based encoder and decoder we were able to generate realistic, original samples., These results demonstrate the ability to realistically augment datasets, improving training of machine learning models designed to predict concentration …

A Mathematical Framework Of Human Thought Process: Rectifying Software Construction Inefficiency And Identifying Characteristic Efficiencies Of Networked Systems Via Problem-Solution Cycle, Jonathan Sarbah-Yalley

Master's Theses

Problem

The lack of a theory to explain human thought process latently affects the general perception of problem solving activities. This present study was to theorize human thought process (HTP) to ascertain in general the effect of problem solving inadequacy on efficiency.

Method

To theorize human thought process (HTP), basic human problem solving activities were investigated through the vein of problem-solution cycle (PSC). The scope of PSC investigation was focused on the inefficiency problem in software construction and latent characteristic efficiencies of a similar networked system. In order to analyze said PSC activities, three mathematical quotients and a messaging wavefunction …

Multi-Directional Warning Message Dissemination Protocol Based On Motion Vector Clustering For Vanets, Hung-Chou Wang

Master's Theses

Problem. Most broadcast suppression protocols in vehicular ad hoc networks (VANET) mainly focus on one-dimensional message dissemination model for both highway and urban scenarios. Due to the non-line-of-sight (NLOS) problem occuring frequently in urban scenario, protocols mostly rely on either infrastructure or the vehicle that is passing through the intersection to forward the message in multiple directions manner. However, these one-dimensional message dissemination models fail to take into account realistic road topologies and traffic distribution. As a result, they tend to miss some possible dissemination directions.

Method. Vehicles travelling on the same road share similar motion pattern due to the …

A Comparison Of Computer-Based Training And Traditional Classroom Training For The Sad Corporation Job Placement Team, Janice Brown Adams

Dissertations

Purpose: The purpose of this research study was to compare the effectiveness of computer-based training with traditional classroom training; both instructional delivery methods utilized the traditional classroom curriculum as a means of delivery for an on-the-job company-sponsored program. The JAW-SAD Human Resource Center (HRC) is a training facility for the SAD Corporation, a major international manufacturing company. The HRC is responsible for conducting continuous training and certification of SAD employees. Historically, training has been provided through conference workshops with a traditional classroom training format. The motivation and importance for this study was influenced by an administrative requirement to determine the …