Skip to content

Latest commit

 

History

History
10 lines (6 loc) · 382 Bytes

README.md

File metadata and controls

10 lines (6 loc) · 382 Bytes

Min-Hashing

Authors: Harshat, Nanthini

  • Getting the signature matrix of the ml-20m dataset with respect to users and movies they have rated >2.5

  • Generating a set representation of the users who like a set of movies

  • Using the signature matrix to efficiently retrieve similar pairs of users

  • Using LSH to retrieve the nearest neighbor (most similar user) to an input user