Relevance-Based Ranking of Video Comments on YouTube

There are many web platforms that are used to share non-textual content, such as videos, images, animations, etc. and that allow users to add comments for each item. YouTube is probably the most popular of them, with billions of videos uploaded by its users, but also with billions of comments for all these videos. While most of the videos only have a couple of comments, the most debated ones have tens of million comments. However, the platform does not provide any mechanism to filter the comments in order to get the most relevant ones, as the only provided ordering is in descending order of the publishing date. We propose a ranking mechanism for these comments that tries to determine the relevance of each comment to the individual video by automatically linking it to relevant web pages with textual information. Three different ranking methods are discussed and their results are presented in order to offer a comparison among them.