Matching Items (1,867)
Filtering by
- Creators: Beethoven, Ludwig van, 1770-1827
- Creators: Computer Science and Engineering Program
Description
This project presents an attempt to recognise single, handwritten alphanumeric characters with a machine learning-based approach. Using a dataset of 131,600 handwritten characters, I trained a custom convolutional neural network (CNN), which achieved a 87.94% accuracy rate, and I applied the ResNet-152–a well-known published model, which achieved a 87.13% accuracy rate, for comparison. A full-stack web application was also developed in conjunction with the model for demonstration. These results signify that the depth of the model was not the main culprit for the systematic failure of recognising some pairs of characters, and that a different approach must be attempted.
ContributorsHo, Timothy (Author) / Menees, Jodi (Thesis director) / Srinivasan, Aravind (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
Businesses and advertising agencies today often promote their products and services through advertisements across various online services such as Google, Meta, Microsoft, and others. Online advertising plays a crucial part in acquiring new clients, resulting in the business's success. However, the interpretation of essential advertising data, such as cost per click and conversion rate, across these advertising platforms presents a challenge to smaller businesses and advertising agencies that manage various client accounts across multiple platforms. This project resolves inefficiencies in digital marketing reporting by delivering a study on the design, development, and evaluation of the Advertising Analytics Dashboard, which provides businesses and agencies such as Vloe, a Quebec-based advertising agency, a unified platform for retrieving tracking analysis and advertising metrics for multiple accounts across multiple online advertising platforms (Vloe, n.d.). Users through the Advertising Analytics Dashboard can detect embedded tracking tags such as Google Analytics and Meta Pixel, and connect their advertising accounts to retrieve data across Google Analytics, Google Ads, Meta Ads, and Microsoft Ads. The platform supports OAuth login, query saving, PDF report generation, and dynamic localization in English and Canadian French. Upon deploying the website, an usability study with 14 participants indicated that users found the dashboard intuitive and the retrieved data accurate. The project successfully delivered and demonstrated how the user-friendly, simplified, and streamlined Advertising Analytics Dashboard provides value to small businesses and advertising agencies for viewing and analyzing tracking and advertising data across multiple platforms in one centralized location.
ContributorsMahajan, Neil (Author) / Chavez Echeagaray, Maria Elena (Thesis director) / Dansereau, Christine (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
This thesis presents the design, implementation, and evaluation of Beacon, a self-hosted error
monitoring platform developed for ASU Online. Beacon addresses the challenges of observability in digital learning environments by providing comprehensive error tracking, real-time alerting, and customizable dashboards while ensuring data sovereignty, cost efficiency, and scalability. Built on an event-driven architecture using open standards, Beacon aims to improve issue resolution times, enhance student experience, and reduce operational costs compared to commercial alternatives. The system leverages OpenTelemetry for data collection, Kafka for event processing, and Elasticsearch for storage, demonstrating how these technologies can be integrated to create a robust observability solution tailored to educational technology needs.
ContributorsJuntilla, Ben (Author) / Menees, Jodi (Thesis director) / Jagannath, Shruthi (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
The Robotics Design Fundamentals Educational VR Experience is a game designed to help players learn and understand the concepts of robot configuration, basic kinematics, and forward kinematics through an embodied approach in virtual reality. This Virtual Reality (VR) experience focuses on reinforcing player's understanding of the Denavit-Hartenberg (DH) parameter derivation process, a key step in developing forward kinematic models. Forward kinematic models are extensively used in the field of robotics, since they act as an intermediate step in the development of more complex models, such as inverse kinematic models, so it is important for students to be able to quickly and confidently derive forward kinematic models. By analyzing VR game design best practices, characteristics of effective embodied learning approaches, and current educational robotics simulators, The Robotics Design Fundamentals Educational VR Experience aims to be an effective tool for students to practice the process of deriving DH parameters and forward kinematic models.
ContributorsJung, Damon (Author) / Johnson, Mina (Thesis director) / LiKamWa, Robert (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / Computing and Informatics Program (Contributor)
Created2025-05
Description
The perfect anti-cheat software for a first person shooter that balances protecting user privacy and effective cheat detection in a modern age where dishonest methods of gameplay are rampant within competitive games. By utilizing the inherent protections servers have against third party attacks, by removing the software off of the client, all of the detection methods are placed in an external area, where cheaters are determined by behavior that is tracked through statistical trackers placed in the game. By measuring multiple key features including Illegal Trace Time, Trigger Time, and Mouse Flick Speed. Each of these measured attributes relate to commonly used cheats in first person shooters, which is the target for this anti-cheat machine learning model. By gathering a wide range of statistics and figuring out the average player’s statistics, it would be possible to determine if a player is using external programs to gain an unfair advantage.
ContributorsKim, James (Author) / Kobayashi, Yoshihiro (Thesis director) / Baek, Jaejong (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05
Description
The following addresses the challenge of effectively connecting mentors with student-run ventures at Arizona State University (ASU). Based on observations and interviews conducted with project sponsor Dr. Byrne and other mentors, the existing informal, referral based approach is inefficient, meaning that potential mentorship opportunities were lost.
To streamline the process, a web platform was developed. This site enables ventures to create structured profiles highlighting concise value propositions and other key indicators, empowering mentors to proactively identify suitable ventures. The technical implementation utilized Next.js for the frontend, Firebase and Firestore for the authentication and storage, and TailwindUI for styling. The result is a user friendly and scalable minimal viable product.
ContributorsMulderink, Matthew (Author) / Osburn, Steven (Thesis director) / Byrne, Jared (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor) / Tech Entrepreneurship & Mgmt (Contributor) / Dean, W.P. Carey School of Business (Contributor)
Created2025-05
Description
Machine learning is defined as: the use and development of computer systems
that are able to learn and adapt without following explicit instructions, by using
algorithms and statistical models to analyze and draw inferences from patterns in
data. Neural Networks are one such system that uses a series of connected nodes
(called neurons) where each connection is adjustable according to certain parameters
called ”weights”. Additionally each neuron has an adjustable bias value which adds
a fixed amount to the sum of the neurons connections. Deep learning is an algorithm
for tuning the parameters (i.e. weights and biases) of a network in order to best fit a
given problem.
For this project the problem I have selected is that of symbol recognition. I am
using the MNIST Handwritten Digit dataset which contains 70,000 images of digits
(0-9). Each image is a 28x28 grid of pixels with values from 0-255. The goal of my
system is to take an image and produce the matching 7 segment display representation
of the number in the image.
The goal of this project was to investigate methods for reducing the cost to complete
this identification task. These methods are separated into three main sections:
1. Topology
2. Knowledge Distillation
3. Network Pruning
The Topology section investigated the impacts of changing the layer sizes of a
network. In this section I found that it is better to have more connections to the
output layer than to any other layer in the network. This makes sense as the output
layer is what we expect to have the results we are looking for and so giving it more
data allows for better differentiation.
The Knowledge Distillation section focused on a training method of the same
name. This method involves the use of a larger, well trained teacher model. This
model is used as an example for a student model to try and mimic. I found that
this setup can work very well, with the student often outperforming the teacher after
the same amount of training. However, the target of the training must be chosen
carefully to avoid interfering with the student’s learning process.
The final section focused on network pruning. Pruning is a process that happens in
biology to remove weak connections to make a neural network more efficient. I found
that automatically removing connections throughout the training process worked exceptionally
well with results of our pruned network matching the control. However, I
did find that more investigation is needed to identify which connections are the most
important before removing them at the start.
ContributorsFrink, Ethan (Author) / Osburn, Steven (Thesis director) / Bazzi, Rida (Committee member) / Barrett, The Honors College (Contributor) / Computer Science and Engineering Program (Contributor)
Created2025-05