/MovieLens-Pt2-Collaborative-Filtering-Recommender-Systems-Graphlab

Used collaborative filtering to produce a movie ratings classifier and recommender. Used the Netflix Prize's top 10 performers to perform benchmarking and placed 6th. Also explored content-filtering models using dimensionality reduced features and k-means, naive bayes, decision trees, and random forests classifiers in separate repo: MovieLens-Pt1-Content-Filtering-Recommender-Systems-Pandas-SKlearn

Primary LanguageJupyter Notebook

MovieLens-Pt2-Collaborative-Filtering-Recommender-Systems-Graphlab

Project Summery:

Used collaborative filtering to produce a movie ratings classifier and recommender. Used the Netflix Prize's top 10 performers to perform benchmarking and placed 6th. Also explored content- filtering models using dimensionality reduced features and k-means, naive bayes, decision trees, and random forests classifiers.

File Structure Summary:

The project's files are organized into the following structure:

Code Folder - In Python - Contains IPython Jupyter notebooks which perform contain all data analysis and any custom functions built in separate python script files. (Currently in progress)

Resources Folder - All the raw and derived data used in the project. Also contains original research describing how the origin of the raw data.

Presentation file - A deck of initial results presented in early 2016.

Project Overview file - Provides a high level overview of the insights made throughout the data analysis. (Currently in progress)

README file - You're reading it. Describes logistics what things are doing and how they are organized.