After my week 8 overview, I chose to play with my home YouTube account. The main purpose of my YouTube account, is to watch Greek movies, listen to Greek music and English pop. Thus, the recommendations in my profile are mainly about Greek music and movies.
As Nidhi Nangia mentions, the browsing history on Google, Chrome and YouTube largely influence the search results of the YouTube recommendations which clearly explains the suggestions in my account.
For that reason, I decided to feed new data to the YouTube algorithm, by choosing rock music; a type of music that I haven’t listened for a long time.
After familiarising myself with the algorithmic theory, a question was raised. Is the frequency of watching a video or the variety of the videos that affect the recommendations? Ideally, I would need to different account to check that.
Also, another question that I wanted to explore is how long it will take to the algorithm to bring back choices from the past. As I used to listen to rock ballads a lot, I wondered if the algorithm has kept this information and started suggesting rock music again after my first couple of videos.
However, as there are a lot of parameters that I need to take into consideration, my algorithm play could only lead me in some assumptions and not clear and in stable results.
