A project to improve the performance of a recommendation system by transforming collaborative filtering into supervised learning and assess impact of adding more features to input
- We used the movie lens dataset
- We have considered multi-criteria such as age, gender etc rather than only considering the ratings
- Singular Value Decomposition was used to convert the sparse vector into a fixed set of features and extract latent variables.
- Artificial neural network for supervised learning was used to extrapolate the unknown variables.