Skip to content
This repository has been archived by the owner on Mar 2, 2020. It is now read-only.

Recommendation System using Collaborative Filtering

License

Notifications You must be signed in to change notification settings

irmowan/Collaborative-Filtering

Repository files navigation

Implementation of Collaborative Filtering

Collaborative Filtering is a technique used by some recommender systems.

This repository is the Python implementation of Collaborative Filtering.

Usage

Run:

> python main.py

Notice:

Python Version: 3.5.1

Required modules: Numpy, Pandas, Matplotlib

Need to download the dataset first and put it in the dataset/ folder.

Or, you can see the result without downloading the dataset.

Dataset

MovieLens, 100K dataset

Report

推荐系统的协同过滤算法实现和浅析 is the pdf version of report.

File tree

Python files:
├── main.py			# Main python file including training and testing.
├── predict.py		# Predict functions.
├── utils.py			# Some useful functions, including calculating.
├── var.py			# Define global variables.

Jupyter Notebook files:
├── Cross Validation.ipynb	# Main file of Collartive Filtering using Cross Validation.
├── TopK.ipynb				# File to choose K in Top-K algorithm.
├── alpha.ipynb				# File to choose alpha in model blending.
├── MovieLens.ipynb			# Early version file for data cleanning.
├── CF.ipynb					# Early version file about Collarative Filtering.

Others:
├── LICENSE				# MIT LICENSE
├── papers				# ignored. Papers have been cited in report.
├── dataset				# ignored. You can get it from GroupLens Website.
│   ├── ml-100k			# 100K MovieLens dataset
├── report
│   ├── Report.tex		# Raw Tex file. Using XeTeX as the engine.
│   ├── Report.bib		# References.
│   ├── Report.pdf		# Exported pdf report.
│   ├── Plot				# Plot folder. Including Echarts.
│   ├── K-figure.png		# Pictures included in the report.
│   ├── alpha-figure.png
│   └── rating-pie.png

License

MIT LICENSE

Author

Irmo

2016.6

About

Recommendation System using Collaborative Filtering

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published