Publication Date

Spring 2023

Degree Type

Master's Project

First Advisor

William Andreopoulos

Second Advisor

Robert Chun

Third Advisor

Nada Attar


NFT (Non-Fungible Token), UI (User Interface), ML (Machine Learning)


NFT Prediction Systems are web applications that provide their users with valuable insights about the artifact. These insights are useful for investors and collectors to make better decisions about their purchases. This project builds upon the same concept of prediction by developing a web application to dynamically provide recommendations based on user input and training an ML model to predict their cost. Preliminary work for the prediction system involved data collection, pre-processing, analysis, and filtering of large datasets from diverse sources. The project focused on the development of a user- friendly UI to enable seamless categorization of search results generated by the Machine Learning model. The ML model serves as the backbone of the prediction system. It is trained using the occurrences of keywords in the NFT description and title alongside other parameters such as price. The model adapts according to the user input and provides an output that can help the user select the appropriate NFT. Experiments were conducted to determine the accuracy of the prediction system by altering the user input and analyzing the resulting outcome.

Available for download on Friday, May 24, 2024