https://github.com/moeh-jama/user-based-predictor
Collaborative filtering algorithm used to predict a users rating for some unseen item using its k closest (similiar) rating users. Built using pearson's correlation, Knn, and resnick algorithm.
https://github.com/moeh-jama/user-based-predictor
collaborative-filtering pearson-correlation predictive-modeling rating-system recommender-system
Last synced: 4 months ago
JSON representation
Collaborative filtering algorithm used to predict a users rating for some unseen item using its k closest (similiar) rating users. Built using pearson's correlation, Knn, and resnick algorithm.
- Host: GitHub
- URL: https://github.com/moeh-jama/user-based-predictor
- Owner: Moeh-Jama
- Created: 2018-12-22T17:25:12.000Z (over 6 years ago)
- Default Branch: master
- Last Pushed: 2021-12-10T11:10:20.000Z (over 3 years ago)
- Last Synced: 2025-01-05T20:13:38.860Z (5 months ago)
- Topics: collaborative-filtering, pearson-correlation, predictive-modeling, rating-system, recommender-system
- Language: Python
- Size: 7.81 KB
- Stars: 0
- Watchers: 0
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# user based predictor
using a csv of ratings & movies that was gotten from https://grouplens.org/datasets/movielens/
collect n-users and their ratings. Purpose of the repo is to find the predictive rating for a user a on an unseen, to user a, item j. Find the k nearest neighbours of user a who have high correalation to them and have previously rated item j. Use the resnick algorithm to then find the predicitive rating for item j for user a.
Main-File: main_script.py