Collaborative Filtering

Implemented an efficient algorithm for collaborative filtering and used it to recommend movies to users based on their ratings.

The Coding

All the code was written in Python without using any of the pre-existing ML libraries such as tensorflow, sklearn, etc. The code can be found on my GitHub. The following are the highlights of the code:

  1. Implemented the collaborative filtering algorithm as described in the original paper by Breese et al.
  2. Evaluated on the Netflix dataset.

The Experiments

The major challenge in implementing the algorithm was the runtime. Only using pyhton and for loops the time taken was approximately a day. I had to use numpy’s efficient vectorization to reduce the runtime. The runtime of the algorithm was reduced to 15 mins.

Total runtime : 15 mins

Mean absolute error = 0.773

RMS error = 2.875

The Learnings

I like to categorize my learnings into two categories: The Profound and The Subtle. The Profound are the key questions that were answered through this project. The Subtle could be an intuition I had, insight I discovered or a clever python hack I learnt from this project. For example, print statement has a parameter called end which can be used to change the default newline character to something else.

The Profound

Q1. What can we say about the mean absolute error ?

The mean absolute error is less which signifies that the algorithm gives a good prediction but the variance is very high which is signified by the high RMS error.

The Subtle

  1. Numpy gives a super fast runtime. The vectorization of the code was very easy. It is a little difficult to get your head aorund the vectorization but once you do, it is very easy to implement.