DESIGN AND DEVELOPMENT OF A MACHINE LEARNING-BASED SENTIMENT ANALYSIS SYSTEM TAILORED FOR YOUTUBE
Abstract
The rapid expansion of information technology and artificial intelligence has transformed the
field of natural language processing. This has however, enabled the development of innovative
applications in sentiment analysis. This project therefore leverages on these advancements in
machine learning to develop a model that is predictive for sentiment analysis on YouTube,
utilizing a dataset of 1,048,576 comments which was sourced from Kaggle a reputable data
source (https://www.kaggle.com/datasets/kazanova/sentiment140?resource=download). The
system that is developed in this work of research employs a Neural Network architecture to
classify comments into positive, negative, or neutral sentiments. The dataset consisted of
1,043,576 training samples (99.52%) and 5,000 testing samples (0.48%). The developed model
however comprises of 4 layers, trained over 10 epochs. The goal of this project is to evaluate the
accuracy of our model in detecting user sentiment. Running locally on the user's machine, our
system has the potential to assist content creators, marketers, and researchers in analyzing user
opinions and preferences on YouTube and also in other.
Full-Text Access Notice
In accordance with the NERD Policy on promoting peer-reviewed publication, public access to the full text of a project, thesis or dissertation is restricted for three years, allowing the author and supervisors sufficient time to pursue peer-reviewed publication.
During this period, researchers with legitimate academic or research purposes may request authorisation directly from the author to enable NERD to release the indexed full texts of the work using the form below.