This is an individual summer research project sponsored by Grinnell College during summer 2019 under the supervision of Prof. Fahmida Hamid. The research has been accepted to the 33rd FLAIRS conference.
This project aims at recommending music artists for Twitter Users based on their music tweeting history. This model takes into consideration users' habit of tweeting, context of each tweet and their implicit preferences (via the frequency of tweets).
To run this program, download all the data.csv.zip files and two source code in Jupyter Notebook. Other csv files can be created by running the "MMTD Data Processing.ipynb". The final are saved in "Artists Recsys with Evaluation all (train-test = 60-40).ipynb". The "Song Recommender System.ipynb" is a program that is still in process.