https://github.com/timvisee/hhs-p7-movie-recommendation-engine

:movie_camera: Big data project for college (HHS) period 7
https://github.com/timvisee/hhs-p7-movie-recommendation-engine

algorithm hadoop recommendation-engine spark

Last synced: 15 days ago
JSON representation

:movie_camera: Big data project for college (HHS) period 7

• Host: GitHub
• URL: https://github.com/timvisee/hhs-p7-movie-recommendation-engine
• Owner: timvisee
• Created: 2017-02-20T13:04:18.000Z (almost 8 years ago)
• Default Branch: master
• Last Pushed: 2017-04-07T15:16:28.000Z (almost 8 years ago)
• Last Synced: 2024-11-15T03:28:25.769Z (3 months ago)
• Topics: algorithm, hadoop, recommendation-engine, spark
• Language: Jupyter Notebook
• Homepage:
• Size: 670 KB
• Stars: 3
• Watchers: 7
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Movie Recommendation Engine

Our try on a movie recommendation engine in Apache Spark using Jupyter Notebook.

Our implementation uses machine learning which is trained by a given data set.

Then, the system is able to predict ratings a user will give on movies s/he

didn't watch yet.

The script consists of two parts. The first part is the dataset parsing and

machine learning training logic.  

The second part loops through all users to suggest the top 10 movies for them

based on the predicted ratings for that user on movies s/he didn't watch yet.

The engine is available in the [MovieRecommendationEngine.ipynb](MovieRecommendationEngine.ipynb)

file. Make sure to put the [required](#requirements) data files in the same

directory as the notebook file.

## Requirements

This project has various requirements:

* Spark with Hadoop, to run the Notebook with the algorithm's code.

* Data files:

    * `movies_full.csv`

    * `movies_small.csv`

    * `ratings_full.csv`

    * `ratings_small.csv`

The data files can be fetched from [movielens](https://grouplens.org/datasets/movielens/).