Finding YouTube's top recommendations from any query
Switch branches/tags
Nothing to show
Clone or download
Guillaume Chaslot
Guillaume Chaslot Update README.
Latest commit f32694d Jan 17, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
data New result files Nov 7, 2016
graph-viz Adding graph visualization in d3js. Jan 9, 2017
results - Enable multiple queries separated by quotes Mar 22, 2017
.gitignore Adding a .gitignore file May 29, 2017
README.md Update README. Jan 17, 2018
follow-youtube-recommendations.py Cleanup. Jan 17, 2018
launch.sh New result files Nov 7, 2016
requirements.txt Adding dep requirements May 29, 2017

README.md

Exploring YouTube recommendations

Where does YouTube's recommendation algorithm bring the user if he follows the recommendations? This robot follows the recommendations on YouTube from a given search query to find out where they bring you. After searching on YouTube and following the recommendations, it prints the 50 most recommended videos.

Installation

To install the project's python dependencies, you can use pip:

pip install -r requirements

Used dependencies:

Example of usage:

python2.7 follow-youtube-recommendations.py --query="global warming,vaccines,nasa" --searches=4 --branch=4 --depth=4 --name="science"

  • --query: query that is made on youtube
  • --searches: number of search results from the query to start with
  • --branch: branching factor = number of recommendations that are followed
  • --name: name under which it will be saved
  • --alltime: add this option if you want to start from the most viewed videos for the query (using the option filter by viewcount on youtube)