Home Courses Netflix Movie recommendation system ML Models:Surprise library

ML Models:Surprise library

Instructor: Applied AI Course Duration: 6 mins

This content is restricted. Please Login

Prev

Next

Computing Similarity matrices:Does movie-movie similarity work?

Overview of the modelling strategy.

Netflix Movie Recommendation System

1.1

Business/Real world problem:Problem definition

6 min

1.2

Objectives and constraints

7 min

1.3

Mapping to an ML problem:Data overview

4 min

1.4

Mapping to an ML problem:ML problem formulation

5 min

1.5

Exploratory Data Analysis:Data preprocessing

7 min

1.6

Exploratory Data Analysis:Temporal Train-Test split.

6 min

1.7

Exploratory Data Analysis:Preliminary data analysis.

15 min

1.8

Exploratory Data Analysis:Sparse matrix representation

8 min

1.9

Exploratory Data Analysis:Average ratings for various slices

8 min

1.10

Exploratory Data Analysis:Cold start problem

5 min

1.11

Computing Similarity matrices:User-User similarity matrix

20 min

1.12

Computing Similarity matrices:Movie-Movie similarity

6 min

1.13

Computing Similarity matrices:Does movie-movie similarity work?

6 min

1.14

ML Models:Surprise library

6 min

1.15

Overview of the modelling strategy.

8 min

1.16

5 min

1.17

Google drive with intermediate files

2 min

1.18

Featurizations for regression.

11 min

1.19

Data transformation for Surprise.

2 min

1.20

Xgboost with 13 features

6 min

1.21

Surprise Baseline model.

9 min

1.22

Xgboost + 13 features +Surprise baseline model

4 min

1.23

Surprise KNN predictors

15 min

1.24

Matrix Factorization models using Surprise

5 min

1.25

SVD ++ with implicit feedback

11 min

1.26

Final models with all features and predictors.

4 min

1.27

Comparison between various models.

4 min

1.28

4 min