Uses for recommendation engines

Building Recommendation Engines with PySpark

Jamen Long

Data Scientist at Nike

blank matrix with users on y axis and movies on X axis

Building Recommendation Engines with PySpark

blank user/movie matrix and two factor matrices

Building Recommendation Engines with PySpark

blank user/movie matrix and two factor matrices

Building Recommendation Engines with PySpark

blank user/movie matrix and two factor matrices

Building Recommendation Engines with PySpark

matrix axes in factor matrices

Building Recommendation Engines with PySpark

factor matrices with latent feature axes

Building Recommendation Engines with PySpark

matrix and factor matrices with values

Building Recommendation Engines with PySpark

high horror movie ratings in matrix

Building Recommendation Engines with PySpark

low drama movie ratings in matrix

Building Recommendation Engines with PySpark

high drama ratings in matrix

Building Recommendation Engines with PySpark

low horror ratings in matrix

Building Recommendation Engines with PySpark

respective scores in factor matrices

Building Recommendation Engines with PySpark

respective scores in factor matrices

Building Recommendation Engines with PySpark

original matrix with factor matrices with values

Building Recommendation Engines with PySpark

factor matrix with movies all scoring high in one latent feature

Building Recommendation Engines with PySpark

four unrelated movies with high scores in one latent feature

Building Recommendation Engines with PySpark

matrix with these movies with high avg scores

Building Recommendation Engines with PySpark

unrelated movies with respective Shakespeare work

Building Recommendation Engines with PySpark

Let's practice!

Building Recommendation Engines with PySpark

Preparing Video For Download...