Project Abstract

A web application has been developed mainly for personal use, to aid data collection and analysis of the YouTube recommendation algorithm.The use of recommender systems has grown enormously over the past few years, with them being an integral part of major companies such as Google, Facebook, Twitter, Netflix, Microsoft and Spotify. YouTube is the second most visited website globally, with monthly traffic of over one-hundred billion users.The project hopes to determine if YouTube watch history is trivial to recommended videos. For example, if the user watches one disturbing video, will they be recommended a series of disturbing videos.The aim of this project is to create a web application that collects data on YouTube video recommendations so that we can analyse the YouTube AI?��s recommendation system.

Keywords: Analysis of Recommendation Algorithm, Web Application, Data Visualization

 Conference Details

Session: Poster Session B at Poster Stand 27

Location: Sir Stanley Clarke Auditorium at Wednesday 8th 09:00 – 12:30

Markers: Muneeb Ahmad, Tom Owen

Course: MEng Computing, Masters 4th Year

Future Plans: I’m looking for work