/item_based_collaborative_filtering

item based collabortive filtering for movie datasets

Primary LanguagePython

item_based_collaborative_filtering

Requirements

  • Python 3.5
  • click ( for progress bar )
  • scikit-learn ( only for saving model )
  • numpy
  • scipy

Installation

$ git clone https://github.com/kuboshizuma/item_based_collaborative_filtering.git
$ pip install -r requirements.txt

Preparation

MovieLens datasets is used in this implementation. For demo you need to download this datasets.

Click the following button below, and you can download datasets soon.

Download datasets

After that, unzip the datasets and put the directory in top directory, naming it datasets.

Documentation

Create model

This command takes a few tens of seconds to complete.

$ python set_model.py

Display recommend items

$ python main.py (user_id)

for example

$ python main.py 24