Music Recommender System Using Apache Spark and Python

A recommender system using Alternating Least Squares matrix factorization technique to recommend new artists to a user based on implicit feedback.

Technology Stack:

• Spark MLlib
• Python
• IPython

Required Installation:

• JDK and JRE
• Python 2.7.x
• Python developer dependencies - sudo aptget install pythondev
• Pip - sudo aptget install python-pip
• IPython - sudo pip install ipython
  sudo pip install jupyter
• Apache Spark - wget http : //apache.arvixe.com/spark/spark-1.5.2/spark-1.5.2-bin-hadoop2.6.tgz
  tar-xzf spark-1.5.2-bin-hadoop2.6.tgz

Instructions for running:

git clone https://github.com/gsrajadh/Music-Recommender-Using-Spark.git
cd Music-Recommender-Using-Spark
IPYTHON_OPTS = "notebook" $SPARK_HOME/bin/pyspark

Dataset:

• A modified subset of the publicly available dataset from Audioscrobbler containing 150k users