Abstract

Cryptocurrency markets are highly volatile and driven by rapidly shifting public sentiment and attention. Traditional financial models, reliant on historical pricing data, often fall short in capturingthereal-time,sentiment-basedbehaviourofcryptoinvestors. Among social platforms, X(formerly Twitter) stands out as a key influencer in crypto discussions, offering a rich source of public sentiment. The thesis presents a machine learning-powered tool that leverages tweet volume and sentiment to analyse short-term cryptocurrency trends. The primary aim is to design a practical analysis system that detects and interprets social media ”hype” around specific cryptocurrencies. By allowing users to choose a cryptocurrency name from a list, the tool provides real-time evaluation of tweet volume and sentiment polarity to estimate potential price direction. These insights are visualised across several days through charts displaying sentiment trends, tweet activity, and corresponding price movement. Data was obtained from Kaggle, containing historical tweets from X. Sentiment analysis was conducted using VADER, with preprocessing to structure data appropriately for classification. While the main focus is detecting hype and predicting directional movement using supervised methods, exploratory ideas such as automatic trend detection without prior input are reserved for future work. The tool successfully predicts directional price movements (up or down) for a specific day, supported by confidence scores. It offers decision-support value to analysts and investors seeking real-time, sentiment-informed crypto insights. Future enhancements include integrating richer datasets via the X API, using advanced sentiment models like BERT, and implementing unsupervised methods to automatically surface emerging cryptocurrencies.

Library of Congress Subject Headings

X (Social networking service)--Data processing; Sentiment analysis; Cryptocurrencies--Prices--Forecasting; Natural language processing (Computer science); Machine learning

Publication Date

5-20-2025

Document Type

Thesis

Student Type

Graduate

Degree Name

Professional Studies (MS)

Department, Program, or Center

Graduate Programs & Research

Advisor

Sanjay Modak

Advisor/Committee Member

Ioannis Karamitsos

Campus

RIT Dubai

Plan Codes

PROFST-MS

Download

Share

COinS