George, Sharon (2023) AI-Enhanced Product Recommendation System using YouTube Comments Analysis. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (2MB) | Preview |
Preview |
PDF (Configuration Manual)
Download (1MB) | Preview |
Abstract
This project’s aims to improve any AI driven product recommendations by analysing the YouTube comments and creating a user friendly Recommendation System. Traditional recommendation systems often struggle to adapt to and understand user sentiments quickly. To overcome this, we use YouTube comments as a primary data source and employ the Convolutional Long Short-Term Memory (CLSTM) algorithm for sentiment analysis. The project is unique because it uses real time YouTube comments to create a dynamic dataset for the recommendation system. The CLSTM algorithm is really good at figuring out feelings in text. It gives a lot of details about what users like and tries to make the product suggestions even better by getting more accurate and personalised. In conclusion this project uses sentiment analysis, artificial intelligence and recommendation systems to make recommendations. The CLSTM algorithm in the project performs well with an accuracy of 89%.
Actions (login required)
![]() |
View Item |